Skip to main content

Full text of "Genetic Studies Of Genius Volume I Mental And Physical Traits Of A Thousand Gifted Children"

See other formats


156.765 G32a v.l 58-00850 

Genetic studies of gejiius 




JAN 9 R 1966 

y,/ ;:- ]-;H-, 



3 i 



GENETIC STUDIES OF GENIUS 



EDITED BY 

LEWIS M. TERMAN 



VOLUME I 



MENTAL AND PHYSICAL TRAITS OF A 
THOUSAND GIFTED CHILDREN 



GENETIC STUDIES OF GENIUS 

Edited by LEWIS M, TERMAN 



Volume I. MENTAL AND PHYSICAL TRAITS OF 
A THOUSAND GIFTED CHILDREN 

By LEWIS M. TERMAN and Others 



Volume II. THE EARLY MENTAL TRAITS OF 
THREE HUNDRED GENIUSES 

By CATHERINE M. Cox and Others 



ACKNOWLEDGMENT 

The investigations reported in these 
volumes were made possible by two appro- 
priations from the Commonwealth Fund, 
supplemented by an appropriation from 
the Thomas Welton Stanford Fund, 



GENETIC STUDIES OF GENIUS 



VOLUME I 

MENTAL AND PHYSICAL TRAITS OF A 
THOUSAND GIFTED CHILDREN 

SECOND EDITION 

LEWIS M. TERMAN 

Assisted by 

BIRD T, BALDWIN 

EDITH BRONSON 

JAMES C. DsVoss 

FLORENCE FULLER 

FLORENCE L. GOODBNOUGH 

TRUMAN LEE KELLEY 

MARGARET LIMA 

HELEN MARSHALL 

ALBERT H. MOORE 

A. S. lUUBENHEIMRR 

G. M. RVCH 

RAYMOND L. WILLOUGHBY 

JENNIE BENSON WYMAN 

DOROTHY HAZELTINB YATBS 



STANFORD UNIVERSITY PRESS 

STANFORD UNIVERSITY, CALIFORNIA 

1926 




POESS 



COPYRIGHT, 1925 AND 1926, BY 
STANFORD UNIVERSITY PRESS 



All Rights Reserved 



Published April 1925 
Second Edition December 1928 



PTNTE IN THE 
UNITED STAT8S OV AM EH 1C A. 



PREFACE 

It should go without saying that a nation's resources of 
intellectual talent are among the most precious it will ever 
have. The origin of genius, the natural laws of its develop- 
ment, and the environmental influences by which it may be 
affected for good or ill, are scientific problems of almost un- 
equaled importance for human welfare. Many philosophers 
and scientists, from Plato and Aristotle to the present day, 
have recognized the truth of this. A number of factors, how- 
ever, have worked together to postpone until our own time 
the inauguration of research in this field. Among these may 
be mentioned the following: (1) the influence of current be- 
liefs, partaking of the nature of superstitions, regarding the 
essential nature of the Great Man, who has commonly been 
regarded by the masses as qualitatively set off from the rest 
of mankind, the product of supernatural causes, and moved 
by forces which are not to be explained by the natural laws 
of human behavior; (2) the widespread belief, hardly less 
superstitious in its origin, that intellectual precocity is patho- 
logical; (3) the vigorous growth of democratic sentiment 
in Western Europe and America during the last few hun- 
dred years, which has necessarily tended to encourage an 
attitude unfavorable to a just appreciation of native individ- 
ual differences in human endowment; and (4) the tardy 
birth of the biological sciences, particularly genetics, psy- 
chology, and education. 

The publication of Galton's Hereditary Genius, in 1869, 
marks the beginning of a new era. Since that date the in- 
terest in individual differences and their causes has grown 
until these promise to become national issues on such prob- 
lems as selective immigration, the evils of differential birth 
rates, special training for the gifted, and the economic re- 
ward of creative talent Both scientific and popular interest 
along these lines has been greatly intensified by recent de- 
velopments in the psychological methods of measuring in- 
telligence, which have furnished conclusive proof that native 
differences in endowment are a universal phenomenon and 
that it is possible to evaluate them. Educators, especially, 
have been quick to appreciate the practical significance of 



vi TRAITS OF GIFTED CHILDREN 

such differences, first for the training of backward and de- 
fective children and more recently for the education of the 
gifted. Twice in the last four years the National Society for 
the Study of Education has devoted a yearbook to the gifted 

child. 1 

The problems of genius are chiefly three: its nature, its 
origin, and its cultivation. This volume is concerned pri- 
marily with the nature of genius, insofar as this is indicated 
by the mental and physical traits of intellectually superior 
children. On the origin of such children it has only a few 
facts of rather general nature to present, for it has thus far 
not been possible to make a thoroughgoing study of the he- 
redity of our subjects. On the education of the gifted it is 
hoped that the data presented throw considerable light, 
since educational procedure to be sound must always be 
based upon an analysis of the raw material with which it 
deals. Before the present investigation was undertaken, no 
large group of gifted children had ever been studied. Our 
positive knowledge of the physical, mental, and personality 
traits of such children has been extremely limited, and until 
this knowledge is available there can be no basis for intelli- 
gent educational procedure. It is hardly too much to say 
that this field at present is the "Darkest Africa" of education, 
To what extent genius can be created or destroyed by right 
or wrong training is entirely unknown. 

The purpose of the present investigation has been, there- 
fore, to determine in what respects the typical gifted child 
differs from the typical child of normal mentality. Data 
have been collected on more than 1,400 children, each of 
whom ranks well within the top one per cent of the unse- 
lected school population of corresponding age. The greater 
part of this report, however, is devoted to 643 such children, 
who constitute a typical group for whom the data at hand 
are most extensive. Less extensive material is reported for 
a second group of 309 subjects (Chapter XIX), making a 
total, in round numbers, of nearly 1,000 giftted subjects for 
whom data have been analyzed. On many points control 
data have been secured for 600 to 800 unseketed chil- 

^Nineteenth Yearbook, Part II, 1920, pp. 125: Classroom Problems In 
the Education of Gifted Children, by T. S. Henry, edited by 0. M. Whip- 
pie. Also, the Twenty-third Yearbook t Part I, 1924, pp, 443; The Educa- 
tion of Gifted Children, edited by G. M, Whipple, 



PREFACE vii 

dren. The aim has been to collect, so far as possible, in- 
formation of objective nature, although it has not seemed 
wise to reject, altogether, methods subject to the influence of 
the personal equation. In the main, however, the conclu- 
sions are based upon well defined experimental procedures 
which can be repeated ad libitum for purposes of verifica- 
tion or refutation. Whatever erroneous conclusions have 
been drawn from the data at hand, and it would be vain to 
hope that such have been altogether avoided, should in time 
be corrected. 

To Miss Florence Goodenough, who served for one year 
as chief field assistant and for two years as chief research 
assistant, the author's indebtedness is very great; also to 
Professor Truman Lee Kelley, for his assistance in the statis- 
tical treatment of data. To the entire staff, whose names 
will be found on the title page of this book, the author ex- 
presses his deep obligations. Last, but far from least, he 
would express his gratitude and thanks for the hearty spirit 
of cooperation which has been almost universally shown by 
the parents, teachers, and school officials. But for their will- 
ing sacrifices of time and labor in the collection of data, the 
investigation would have been entirely impossible. 

In Volume II Dr. Catharine M. Cox sets forth the results 
of a parallel study of The Early Mental Traits of Three 
Hundred Geniuses. It is believed that the two volumes will 
yield many interesting and instructive comparisons. 

Two other publications should be mentioned which sup- 
plement these studies. They are: (1) Children's Reading; 
A Guide for Parents and Teachers, by Lewis M, Terman and 
Margaret Lima (Appleton, 1925); (2) An Experimental 
Study of Some of the Behavior Traits of the Potentially 
Delinquent Boy, by Dr. A. S. Raubenheimer (Psychological 
Monographs, 1925). These, like the present volumes, were 
made possible by appropriations from the Commonwealth 
Fund. 

LEWIS M. TEHMAN 
STANFORD UNIVERSITY 
January 15, 1925 



PREFACE TO THE SECOND EDITION 

It was hardly expected that the first edition of this study 
would be exhausted within a year and a half of its publica- 
tion. That a second edition Is called for at this time is en- 
couraging evidence of a widespread and growing interest in 
the educational and social problems relating to the con- 
servation of human talent. I am glad to avail myself of this 
opportunity to express my appreciation of the reception 
which Volume I has been accorded and of the interest which 
has been shown in the projected series as a whole. 

The most important changes which have been made in 
this edition have to do with the interpretation of data pre- 
sented in Chapter VI on the size of families from which our 
gifted children come. In the first edition, unfortunately* the 
treatment on this point was at fault because of neglect to 
take account of childless marriages and of celibacy. The re- 
sult was a material overestimate of the fertility of our gifted 
families. The revised treatment shows that the ferility index 
of that stratum of the California population with which we 
are here concerned has decreased by 50 per cent within a 
single generation and that it is at present far below the 
figure which would permit the maintenance of the stock. 

Throughout the text numerous minor changes have been 
made, chiefly of typographical nature* I am indebted to the 
reorganized Stanford University Press for many such typo- 
graphical rearrangements which have improved the ap- 
pearance of the book, 

LEWIS 1C 

STANFORD UNIVERSITY 
December 1, 1926 



vili 



CONTENTS 

CHAPTER PAGE 

I, HISTORY AND DESCRIPTION OF THE INVESTIGATION ... 1 
Binet's work, 2. Early studies of gifted children at Stan- 
ford, 3. The present study, 5. Data collected under first 
Commonwealth grant, 7, Data collected under second 
grant, 12. Assistants, 15. 

II. METHODS AND RESULTS OF THE SEARCH FOR SUBJECTS . . 19 
The problem of search, 19. Selection of main group, 21. 
Nomination blank, 21, Abbreviated Binet test, 23. Use of 
vocabulary test, 25* Methods in primary grades, 26. Cases 
missed, 27. School population canvassed, 29, Analysis of 
sources of information which yielded the main experi- 
mental group, 30. Special ability cases, 34. Search in out- 
side schools, 36. 

IIL COMPOSITION OF THE GIFTED GROUP ........ 39 

Classification, 39. Composition of the main experimental 
group, 40, Distribution of IQ*s, 41. Correction of IQ for 
older cases, 42. Distribution of corrected IQ's, 44. IQ's of 
outside Binet group, 46. Composition of special ability 
group, 47. Scores of outside high school cases, 48. Sex 
ratio, 49. Effect of method of selection on sex ratio, 50. 
Sex ratio in families of gifted, 51. Differential death rate 
of embryos as possible cause of sex ratio among gifted, 
52. Sex difference in variability, 53. 

IV. RACIAL AND SOCIAL ORIGIN 55 

Racial origin of main gifted group, 55. Percentage of 
parentage of gifted, 58. Birthplace of foreign born parents 
of gifted, 59. Birthplace of American born parents and 
grandparents, 60. Occupation of fathers, 61. Comparison 
with oceupaational origin of genius groups, 64. Barr Scale 
ratings of occupational status, 66. Economic status, 72. 
Home ratings on the Whittier scale for home grading, 73, 
Divorce and separation of parents, 74. Neighborhood rat- 
ings, 75. School reports on the home environment, 76. 
Paid employment, 77. Education of parents and grand- 
parents, 78. Size of home library, 81. Summary, 82. 

V. INTELLECTUALLY SWERIOH RELATIVES ....... 85 

Information schedule used, 85. Siblings in the main 
group, 89. Relatives in the Hall of Fame, 91. Relatives 
in Who"* Who, 92. Other relatives of distinction, 96. 
Positions held by parents and grandparents, 101* Dis- 
tinguished relatives of three gifted sibs, 103. Gifted off- 
spring of a Japanese-American marriage, 107. The R 

family, 108. The B family, 109. Summary, HO. 

lac 



x TRAITS OF GIFTED CHILDREN 

CHAPTER PAGE 

VL VITAL STATISTICS ..113 

Size of families, 113. Size of family and education of 
parents, 117. Size of family and degree of superiority of 
children, 117. Miscarriages, 118. Infant mortality, 119. 
Age of parents at birth of gifted child, 119. Order of 
birth, 121. Mortality statistics regarding parents, 125. 
Chronic illnesses of parents, 127. Longevity of grand- 
parents, 128. Tuberculosis among relatives, 130. Heredi- 
tary diseases and defects, 131. Summary, 133. 

VII. ANTHROPOMETRIC MEASUREMENTS, by Bird T. Baldwin . 135 
Method of work, 135. Subjects measured, 136. List of 
measurements, 137. Location of measurements, 138. 
Instruments, 139. Cooperation of parents, 139. Report to 
parents, 142. Unselected groups, 144. Deviations from the 
norm in weight and breathing capacity, 146. Results of 
physical measurements, 148. Sex and age differences, 149. 
Increments of growth, 149. Variability, 152. Indices of 
growth, 152. Correlations of physical traits, 153. Com- 
parisons with control group, 155. Resemblance between 
parents and children, 158. Growth and nutritional 
status, 158. Application of Dreyer*s standard of weight, 
159. Von Pirquet's pelidisi, 160. Development of carpal 
bones, 162. Stages of physiological maturation, 163. 
Nationality, 166. Mental and physical status, 167. Sum- 
mary, 169. 

VIIL HEALTH AND PHYSICAL HISTORY ......... 173 

Schedules used, 173. Gestation period, 178. Weight at 
birth, 179. Conditions of birth, 181. Infant feeding, 182. 
Early health, 185. Early development (walking, talking, 
etc.), 185, Disease history, 187. Accidents and operations, 
191. Miscellaneous defects, 192 ff. Nervous disturbances, 
198. Habitus, 201. Eating habits, 202. Personal hygiene, 
203. Sexual development of boys, 205. Sexual develop- 
ment of girls, 207. Summary, 210. 

IX. MEDICAL EXAMINATIONS, by Dr. Moore and Dr. Bronson . 215 

Subjects examined, 215. Examination schedule arid pro- 
cedure, 217. General impression, 219. Metabolism tests, 
220. Skin, 222. Head, 224. Hearing, 224. Vision, 225* 
Mouth, 227. Dentition, 228. Pharynx and tonsils, 230, 
Condition of thyroid, 231. Cervical glands, 232. Chest 
deformities, 233. Respiration rate, 234. Lung conditions, 
235. Heart and pulse, 235 f. D'Espine sign, 237. Abdomen, 
238. Genitals, 239. Deformities, 240. Neurologic condi- 
tions, 243. Endocrine symptoms, 244. Condition of urine, 
244. Hemoglobin, 246. Blood pressure, 246, Summary, 240. 

X. SCHOOL PBOGRESS AND EDUCATIONAL HISTOHY . . , * 253 

Age-^rade status, 253. School progress quotient, 257, 
Ratings of school work, 258. Evenness of ability, 264* 
Educational history, 266. Attendance, 267. Liking for 
school, 269. Early reading, 271. Grades skipped, 272. 
Forced culture, 273. Hours of home study, 274. Private 
tutoring, 275. Early indications of superior ability, 277, 
Methods of home training, 283* Summary, 285* 



CONTENTS xi 

CHAPTER PAGE 

XI. TESTS OF SCHOOL ACCOMPLISHMENT AND GENERAL 

INFORMATION . , 289 

The Stanford Achievement Tests, 289. Mean achievement 
scores by age, 292. Mean subject quotients, 294. General 
information tests, 296, Mean information scores, 300. 
Mean information quotients, 303. Influence of attendance 
upon educational accomplishment, 304. Summary, 305. 

XII. SPECIALIZATION OF THE ABILITIES OF GIFTED CHILDREN, 

by James C. De Voss , . 307 

The problem, 307. Method of treatment of data, 308. 
Comparison of a gifted and control group by the use of 
Kelley's ratio, 310. Further comparison of type groups 
of gifted and unselected children, 327. Conclusions from 
the two methods of comparison, 339. Nature of the un- 
evenness in the abilities of gifted children, 340. Achieve- 
ment levels and intelligence levels, 345, Case studies, 349. 
Summary, 361. 

XIIL SCHOLASTIC, OCCUPATIONAL, AND OTHER INTERESTS . . 363 
Scholastic interests, 363. Liked and disliked subjects, 368. 
**Easy wf subjects, 370. Occupational interests, 373. Barr 
Scale ratings of preferred occupations, 376. Preference 
for various types of activity, 377. Collections, 379. Sum- 
mary, 382. 

XIV. PLAY INTERESTS, KNOWLEDGE, AND PRACTICE . . . 385 

Description of the test of play interests, 385. Groups 
tested, 393. Derivation of preference indices for ninety 
activities, 393. Masculinity indices of the ninety activi- 
ties, 407. Masculinity ratings of the children, 410. Ma- 
turity indices of activities, 413, Maturity ratings of the 
children, 420. Sociability and activity ratings of the chil- 
dren, 420. Experience and skills, 425, Play information, 
428. Home and school data on play life, 429. Summary, 
437. 

XV. READING INTERESTS 441 

Amount of reading as estimated by parents, 441; by 
teachers, 443. Record of two months* reading, 444, Influ- 
ence of Intelligence on type and range of reading, 447. 
Sex differences in reading interests, 448. Preference rat- 
ings of books read, 450. Favorite books, 451. Summary, 
453. 

XVI. TESTS OF INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS, 

by Jennie Benson Wyman ,..,.., 455 
Introductory statement, 465. Method of approach, 456. 
Preliminary experiment, 457. Derivation of scoring meth- 
od, 461. Reliability of the test, 467. Validity of the test, 
468. Case comparisons, 470, Comparison of gifted and 
control groups, 474, The influence of interest upon school 
achievement, 479. Summary, 482, 



xii TRAITS OF GIFTED CHILDREN 

CHAPTEB 

XVII. TESTS OF CHARACTER AND PERSONALITY TRAITS .... 485 
Earlier work of Voelker and Cady, 485. Description of 
Raubenheimer's tests, 488. The trustworthiness test, 497. 
The Woodworth-Cady questionnaire, 500. Subjects tested, 
505. Derivation of total score, 506. Comparison of gifted 
and control groups, 508, Summary, 516. 

XVIII. TRAIT RATINGS 519 

Earlier trait ratings of gifted, 519. The graphic rating 
scale used, 523. Mean ratings by age, sex, and intelligence 
on twenty-five traits, 531. Mean ratings for ages com- 
bined, 537. The classification of traits, 538. Influence of 
the "generosity" factor in trait ratings, 545. Overlapping 
of gifted and control on personality traits, 546. Hank 
orders of the traits for various groups, 547. Unevenness 
of ratings for the individual children, 549. Certainty of 
judgments, 550. Relative variability of the sexes in trait 
ratings, 553. Reliability of the ratings, 553. Summary, 554, 

XIX. SUMMARY OF DATA ON 309 GIFTED HIGH SCHOOL PUPILS . 557 
The data, 557. Intellectual composition of the group, 558. 
Sex proportions, 560. Geographical location of pupils, 561. 
Ancestry, 563. Birthplace of parents and grandparents, 
565. Occupational background, 568. Health and physical 
data, 570. Sexual maturity, 577. Educational history, 578. 
Tests of general information, 582, Scholastic preferences, 
585. Vocational preferences, 590. Summary, 592. 

XX, Two YEARS AFTER 597 

Annual report form, 597. Age-grade status, 601. School 
progress in reference to recommended promotions, 608. 
School progress and scholarship rating, 611. School prog- 
ress in relation to liking for school, 613. School progress 
and social adaptability, 613. Deportment, application, and 
nervousness in relation to school progress, 619. Changes 
in responsiveness to discipline, 620. Changes in interests 
and in social and character traits, 623. Changes in inter- 
ests as related to changes in social traits, 627. Summary, 
629* 

XXL CONCLUSIONS AND PROBLEMS 631 

Variability of the group studied, 631. Representative 
character of the group, 632. Limitations of data, 632. 
Verifiability of conclusions, 633. The distribution of 
intelligence, 633. Sex ratio, 634, Growth and health, 634. 
The environment hypothesis, 634. School progress, 636, 
Specialization of ability, 636. Interests, 637. Character and 
personality traits, 638. Data for the control groups, 639. 
Prediction of success, 640. Problems, 641. 

INDEX 



LIST OF FIGURES 

rttit'HE PAGE 

1. IQ Distributions for 999 Gifted and 905 Unselected 

Children 47 

2. Location of Physical Measurements . . 138 

3. Individual Growth Curves of Two Sibs, Brother 

and Sister .......... 140 

4. Percentage Deviations of Gifted Children from the 

Baldwin-Wood \Veight-Height-Age Norms ..... 147 

5. Percentage Deviations of Gifted Children from the Norms 

in Breathing Capacity for Age and Height 147 

6. Percentage Deviations of Gifted Children from the Norms 

in Breathing Capacity for Age and Weight 148 

7. Age Curves for Sixteen Physical Traits, Gifted Boys . . 150 

8. Same, Gifted Girls 151 

9. Growth and Development of Gifted and Unselected 

Children 213 

10. Physical Defects in Gifted and Control Groups .... 214 
lOa Subject Quotient Profile of Gifted Children 293 

11. Showing Distributions of Found and True Differences 

Between Scores 318 

12. Profiles of Three Groups of "Good Readers" 332 

13. Profiles of Three Groups of "Good Calculators" 336 

14. Profiles of Three Groups of "Good Spellers" 338 

15. Profile of A. K 341 

16. Profile of H. M. J. . . . 350 

17. Profile of N. M 351 

18. Profile of S. M. F 353 

19. ProfiieofS*D 354 

20. Profile of A. A, .356 

21. Profile of G. F. 357 

22. ProfileofZ. J. 358 

23. Profile of D. E. B 359 

24. Profile of J. D. 360 

25* Men Masculinity Ratings by Age, Gifted and Control 

Groups . 414 

26. Mean Maturity Ratings by Age, Gifted and Control Groups . 423 

27. Mean Scores in Interest Tests, Gifted and Control 

Groups, by Age . * . 477 

28* Circles Used in Trustworthiness Test ........ 498 

29. Squares Used in Trustworthiness Test 499 

30* Mean Total Scores of Gifted and Control Children, 

by Age, on Seven Character Traits 515 

31. Mean Trait Ratings by Age 531-536 

32. Mean Trait Ratings, Ages Combined 539-541 

33. Age Means for Each Class of Traits 543-544 

34. Birthplace of Parents and Grandparents of 

High School Gifted Group 565 

35. Occupational Background Indices and Barr Ratings 

of Preferred Occupations 570 

xiii 



MENTAL AND PHYSICAL TRAITS OF A 
THOUSAND GIFTED CHILDREN 

CHAPTER I 

HISTORY AND DESCRIPTION OF THE INVESTIGATION 

PBELIMINARY EXPLORATION 

This research may be said to have had its beginning 
during the years of the writer's graduate study, 1902-1905. 
He first became interested in the psychology of genius in a 
study of leadership which he made in 1902-1903 under the 
direction of Professor E, H. Lindley, of Indiana University. 1 
While a student at Clark University he reviewed in 1903- 
1904 the medical-psychological literature on precocious 
children 2 and the following year carried out as a doctor's 
dissertation an experimental study of some of the mental 
processes of seven bright and seven dull boys. 3 

However slight the positive contribution of these studies, 
they at least introduced their author to the literature on 
the psychology of genius and gave a keen realization of the 
fact that the field was a promising one for experimental in- 
vestigation. When in 1910 it became possible for the writer 
to return to the problem, the progress which Binet and 
others had made in the field of mentality testing had created 
an entirely new situation. For certain ages, at least, it was 
at last possible to determine with some degree of approxi- 
mation the brightness of a given child, compared with that 
of unselected children of his own age. 

The importance of Binefs work for later studies of in- 
telligence can hardly be overestimated. It has not yet re- 
ceived and possibly may never receive from psychologists 

*A Preliminary Study in the Psychology and Pedagogy of Leadership. 
Pedagogical Seminary, 1904, Vol. 11, pp. 413-451. 

*A Study in Precocity and Prematuration, American Journal of Psy- 
chology, 1905, Vol. 16, pp. 145-163, 

"Genius and Stupidity: A Study of Some of the Intellectual Processes 
of Seven "Bright" and Seven "Stupid** Boys. Pedagogical Seminary, 1906, 
VoL 13, pp 3Q7--373. 



2 TRAITS OF GIFTED CHILDREN 

the appreciation which it deserves. Critical ability, unfor- 
tunately, is far more common than ability to create, and to 
the critical psychologist the imperfections and crudities of 
Binet's methods, both in their practical and in their theoreti- 
cal aspects, have often been more evident than their remark- 
able originality. More than anyone else, it was Binet who 
taught us where to search among mental functions for sig- 
nificant intellectual differences. It was he who gave us our 
first successful intelligence scale and demonstrated the ac- 
tuality of an age development through successive ^hier- 
archies of intelligences." That the term "mental age" which 
resulted from the latter concept has often been misinter- 
preted and misused, does not detract from the importance 
of his contribution. The fact is that previous to the publica- 
tion of Binet's 1908 scale the significance of age differences in 
intelligence was very little understood. Psychologists were 
not aware of the extraordinary and detailed similarity that 
may exist between a dull child of twelve years and a nor- 
mal average child of eight. No one recognized the signifi- 
cance, for future mental development, of a given degree of 
retardation or acceleration. As one who had worked experi- 
mentally upon the diagnosis of intellectual differences in the 
pre-scale period, the present writer had perhaps more rea- 
son than most psychologists to appreciate the value of Binet's 
contribution. He is willing to admit that after spending 
four or five hours a day for several months in adminis- 
tering an extended series of well-selected intelligence tests 
individually to fourteen boys, he was unable, notwithstand- 
ing the large individual differences in performance which 
these tests clearly revealed, to render a judgment as to the 
prognostic significance of the differences found. By the Bi- 
net scale it would have been possible to make a more mean- 
ingful diagnosis after a one-hour test of each child; and it 
would now even be possible to do so after a single hour 
spent in testing the fourteen boys by a group test This ad- 
vance is due (1), to the demonstrated validity of the con- 
cepts of mental "retardation" and mental "acceleration;" 
and (2), to the convenient and readily comprehensible meth- 
od suggested by Binet for evaluating degrees of retardation 
and acceleration in terms of normal mental age units. Pre- 
vious to 1908 it was impossible for any psychologist, after 
devoting any amount of time to intelligence tests of ten or 



HISTORY AND DESCRIPTION 3 

twenty children of different ages, to make a valid compari- 
son of the intellectual abilities found. This is now possible 
for even a well-trained normal school graduate. 

The value of the Binet method in the identification of 
the intellectually gifted became immediately evident to the 
writer when with Mr. H. G. Childs he made trial of the 1908 
scale. It was obvious that children who showed marked ac- 
celeration in mental age were, by any reasonable criterion, 
brighter than children who tested at or below their chron- 
ological age. A little later Stern's suggestion looking toward 
the use of an intelligence ratio, or quotient, refined still fur- 
ther the method of Binet and made possible more accurate 
comparisons of children of different ages. 1 

In 1911 more or less systematic work was begun at Stan- 
ford University in the collection of data on children who had 
made exceptionally high scores in a mental test. In 1913- 
1914 three schools in San Francisco were sifted for bright 
children, and in 1915 certain data were published on 31 
cases testing above 125 intelligence quotient (IQ). a Ratings 
on several traits were secured from the teachers, who also 
filled out a brief information schedule for each child. Some 
of the results of this explorative study were out of line with 
the writer's expectations and in contradiction to earlier 
views which he had published on the supposed evils of pre- 
cocity** It was obvious that these children did not, as a 
group, possess the traits which had been popularly supposed 
to characterize intellectually precocious children, such as 
sickliness, eccentricity, one-sidedness, and lack of social 
adaptability. In passing it may be noted that one of the 
bright children tested in 1911 has taken his Ph.D. degree and 
is (1924) an instructor in a great western university; thatan- 
other has just completed his work for the degree of Sc.D.; 
and that another is studying in the universities of Europe. 

In 1916 the methods used were considerably revised. 
The teacher's information schedule was enlarged, a similar 
information schedule was prepared for the parent to fill 
out, and ratings on twenty traits were secured both from 

*W. Stern: Die psychologischen Methoden der Intellsgenzpriifungen und 
deren Anwendung an Schtilkindern. Sonderabdr. aus Bericht tiber den V 
Kongress f. experimentelle Psychologie, Leipzig, 1912. 

"Lewis M. Terman: The Mental Hygiene of Exceptional Children. 
Pedagogical Seminary, 1915, Vol. 22, p. 534 ff. 

American Journal of Psychology t 1905, Vol. 16, pp. 145-183. 



4 TRAITS OF GIFTED CHILDREN 

parents and from teachers. With the assistance of Margaret 
Hopwood Hubbard, data were collected on 59 cases, most of 
whom had an intelligence quotient above 140. The main re- 
sults of this study have been published elsewhere and need 
not be summarized here. 1 The writer's tentative conclusions 
of 1915 were fully supported. 

The establishment by Stanford University of a research 
fellowship for the study of gifted children, in 1919, was the 
occasion for further revision of method and a stimulus to 
renewed search for cases. The information schedules were 
materially improved and an interest blank was arranged 
for the child to fill out. By the spring of 1921 approximately 
150 cases testing for the most part above 140 IQ had been 
located, and for 121 considerable supplementary data had 
been secured. The results for these 121 cases have not been 
published, but it may be stated that they suggested the fol- 
lowing tentative conclusions: 

1. There is probably a somewhat higher incidence of in- 
tellectual superiority among boys than among girls, 

2. In physical growth and general health gifted children 
as a group excel unselected children of the same age. 

3. Gifted children who attend school are on the average 
accelerated about a year and a half, compared with unse- 
lected children, but on an average they are about two grades 
below that which corresponds to their mental development. 

4. Only a very small minority of intellectually gifted chil- 
dren have been subjected to forced culture or otherwise 
"pushed" in their development. 

5. Heredity is superior. Fifty per cent of the fathers be- 
long to the professional groups; not one to the unskilled 
group. 

6. There is an apparent excess of Jewish cases and a 
deficiency of cases from the Italian, Portuguese, and Mexi- 
can groups living in the vicinity of Stanford University, 

7. Trait ratings and social data give no evidence that 
gifted children tend more often than others to be lacking 
in social adaptability or leadership. However, they are prob- 
ably less superior in social, emotional, and psychophyslcal 
traits than in intellectual and volitional traits. 

l Lewis M. Terman: The Intelligence of School Children. Hougfeton 
Mifflin Company, 1919. See pp. 165-267. 



HISTORY AND DESCRIPTION 5 

During the academic year 1920-21, Mrs. Jessie Chase 
Fen ton served as full-time assistant on the gifted children 
fellowship. Her services did much to lay the foundation for 
the more extensive investigation which was to follow. Be- 
sides collecting considerable data on the social traits of a 
group of 100 intellectually superior children, she assisted in 
the preparation of a report on a gifted young poet, 1 and in 
a summary of recent literature on genius. 2 

Perhaps the most valuable thing gained from the work 
to this point was the experience. Intimate acquaintance 
with a considerable number of gifted children had shown 
the need of certain kinds of home and school data, and suc- 
cessive revisions of information schedules for the use of 
parents and teachers had shown what methods were likely 
to be most dependable in gathering such data. As for con- 
clusions having a statistical basis, none could be established 
except on a far larger number of cases. This is especially 
true of comparisons involving age, race, school grade, oc- 
cupational class, etc. It was clear that for such purposes it 
would be necessary to locate 500 or 1,000 cases by a method 
which would insure that the group selected would be rea- 
sonably representative of intellectually gifted children. 

THE PRESENT STUDY 

The task of locating the desired number of cases and of 
securing the necessary tests and supplementary data was of 
course far too costly to be financed out of the ordinary bud- 
get of a university department. Fortunately, early in 1921 
the directors of the Commonwealth Fund made a grant of 
120,300 to Stanford University to continue and extend the 
research. The purposes of the grant as indicated in the for- 
mal application which the author submitted under date of 
February 23, 1921, were as follows: (1) to increase the num- 
ber of gifted subjects to approximately 1,000; (2) to secure 
at least two intelligence tests of each subject; (3) to secure 
measures of school achievement in at least four or five of 

*Lewis M, Terman and Jessie C. Fenton; Preliminary Report QH a 
Gifted Juvenile Author. Journal of Applied Psychology, 1921, Vol. 5, pp. 
163-178. 

Lewis M. Terman and Jessie M. Chase: The Psychology, Biology, and 
Pedagogy of Genius. Psychological Bulletin, Vol. 17, No. 12, Dec,, 1920, 
pp. 397-409. 



6 TRAITS OF GIFTED CHILDREN 

the school subjects; (4) In the case of a small number of 
cases to give tests of specialized ability; (5) revision of the 
methods of securing trait ratings and social data; and (6) 
follow-up of the subjects for a period of at least ten years. 1 
In 1922, before the end of the first year's work, an additional 
grant of 114,000 was received from the Commonwealth Fund 
for the purpose of extending the study along medical, an- 
thropometric, and psychological lines. This sum was sup- 
plemented by a contribution of $8,000 in money and f 6,000 
in services from Stanford University, The money cost of 
the study here reported, apart from services contributed, 
was therefore $42,300. The contribution of services by the 
University has exceeded the amount stipulated and would 
bring the total cost of the study to more than $50,000. 

The second Commonwealth grant made it possible to se- 
cure for our main group of subjects an thropometric mea- 
surements, medical examinations, character and personality 
tests, and interest tests; and, in addition, to carry out a par- 
allel biographical study of the early mental traits of three 
hundred men and women of genius. 

The first grant was made available in May, 1921, May, 
June, and July were devoted by the writer to the preparation 
of plans, tests, and information blanks and to securing the 
necessary help. The research staff to begin with was as 
follows : 

Assistant Director: 

Dr. T. L. Kelley, Stanford University/ 

Field Assistants: 

Florence Fuller, M.A., University of Minnesota. 
Florence Goodenough, M.A., Columbia University, 
Helen Marshall, MA., Ohio State University. 
Dorothy H. Yates, Ph.D., University of California. 

Office Assistant : 

G. M. Ruch, Ph.D., Stanford University. 

*It was stipulated that the follows p work should be financed by 
Stanford University. 

3 Before the investigation had proceeded far Dr. Kelley left for Eu- 
rope, but after his return, nine months later, he gave invaluable assist- 
ance in connection with the treatment of data. His relation to the entire 
investigation may best be characterized as that of statistical consultant, 
although he should not be held responsible for the statistical shortcomings 
of the report. 



HISTORY AND DESCRIPTION 7 

It will be evident that the success of an undertaking of 
the kind here described depends in no small measure upon 
the qualifications of the field assistants secured. In the 
search for suitable assistants the leading universities of the 
country, were canvassed by the writer in person. Every se- 
lection made proved to be a happy one. Dr. Yates had re- 
cently completed a Ph.D. dissertation on gifted high school 
pupils; Miss Goodenough had worked extensively in mental 
tests and clinical methods with Dr. Leta S. Hollingsworth ; 
Miss Marshall had worked with Dr. Rudolf Pintner in mental 
surveys of school children; and Miss Fuller had assisted 
Dr. M. E. Haggerty for a year in a survey of gifted children 
in Minneapolis. All had had extensive training in the use of 
tests, all had taught in public schools, and all were espe- 
cially interested in the proposed investigation. The assist- 
ance of Dr. Ruch in the work carried on at the University 
was extremely valuable. During 1921-1922 this had to do 
largely with the preparation of tests, especially the achieve- 
ment and general information tests. 

On August 8th the four field assistants began a course of 
five weeks of intensive training at Stanford University in 
preparation for their year's work. Professor L. L. Burlin- 
game of the Department of Biology, Stanford University, 
gave instruction on heredity; Dr. J. Harold Williams, Di- 
rector of the California Bureau of Juvenile Research, on 
methods of collecting field data; Dr. Maud Merrill, instruc- 
tor in psychology, Stanford University, on Binet test pro- 
cedure; the writer, on the literature of genius. Dr. Ruch 
assisted in shaping the plans, and in the preparation of in- 
formation schedules and a general information test. 

The data to be collected for each child chosen for study 
included the following: 

1. Two intelligence tests (Stanford-Binet and Nation- 
al B) 

2. A two-hour educational test (The Stanford Achieve- 
ment Test) 

3. A fifty-minute test of general information in science, 
history, literature, and the arts 

4 A fifty-minute test of knowledge of and interest in 
plays, games, and amusements 

5, A four-page interest blank to be filled out by the 
children 



8 TRAITS OF GIFTED CHILDREN 

6. A two-months reading record to be kept by the chil- 
dren 

7. A sixteen-page Home Information Blank, to be filled 
out by the parents, including ratings on twenty-five traits 

8. An eight-page School Information Blank to be filled 
out by the teachers, including ratings on the same twenty- 
five traits as were rated by the parents 

9. When possible, ratings of the home on the Whittier 
Scale for home grading. 

Field work began early in September, 1921. Miss Good- 
enough and Miss Fuller were assigned to Los Angeles, Miss 
Marshall to San Francisco, and Dr. Yates to Oakland and 
Berkeley. It was thought that the four assistants could can- 
vas grades 1 to 8 in the cities just named and probably also 
in some of the smaller cities. However, the task proved 
more tedious than had been foreseen. After a conference 
with the field assistants late in November, grades 1 and 2 
were eliminated from the formal survey in order that grades 
3 to 8 might be more thoroughly covered. February 1, 1922, 
Miss Bessie Fuller was added to the Los Angeles staff ; Dur- 
ing April, May, and June Miss Marshall was assisted in San 
Francisco by Miss Elizabeth Kellam, and Dr. Yates in Oak- 
land and Berkeley by Miss Beatrice Lantz. Practically the 
entire cities of Los Angeles, San Francisco, and Oakland 
were canvassed in this way, the larger part of Berkeley, and 
part of Alameda. 

With the help of volunteer assistants the following cities 
were covered more or less thoroughly by the same method 
as that used in the larger cities: Santa Barbara (grades 2 
to 6), by Dr. James L. Stockton, of the Santa Barbara State 
Teachers' College; Fresno, by Miss Blanche Cummings, Di- 
rector of Research in the Fresno public schools; San Jose, 
by Professor J. C. DeVoss, San Jose State Teachers* College; 
Santa Ana, by Miss Bess Henry, Director of Research in the 
Santa Ana public schools. With the help of local assistants 
Pasadena, Redlands, Santa Rosa, Palo Alto, Burlingame, 
Kelseyville, Irwin, Sebastopol, Burbank, San Mateo, San 
Bernardino, and a few other cities were canvassed somewhat 
less thoroughly by a simpler method, to be described later. 
In the same way the rural schools of San Bernardino County 
were also covered. 



HISTORY AND DESCRIPTION 9 

Through the cooperation of the high school principals 
of the state a modified form of survey was carried out in 
95 high schools, enrolling approximately 70,000 pupils. The 
regular field assistants canvassed most of the high school 
pupils in Los Angeles, San Francisco, and Oakland. To the 
many volunteer helpers who contributed so much to the 
success of the study as a whole, the writer wishes to express 
his very great obligation. 

The procedure of the regular field assistants was as fol- 
lows: (1) an entire school was canvassed by methods to be 
described in Chapter II; (2) Home Information Blanks were 
distributed at once to the parents of the children selected 
for study, and (in about 95 per cent of the cases) the homes 
of the children were visited and rated and parents were in- 
terviewed; (3) School Information Blanks were distributed 
to the teachers of the children selected; (4) shortly before 
the close of the school year, in May and June, 1922, the se- 
lected cases of a particular city were called together in 
groups of 10 to 50 for the achievement tests, the test on 
plays, games, and amusements, and for the collection of data 
called for in the Interest Blank. At the same time record 
booklets were distributed, in which the children were asked 
to record their reading for a period of two months. 

The data thus collected, together with similar data for 
a control group of about 600 on the plays, games, and amuse- 
ment test, the Interest Blank, the School of Information 
Blank, the Information Test, and the reading records, were 
scored and tabulated in the summer of 1922. In this work 
the writer was assisted by Miss Fuller, Miss Marshall, Miss 
Goodenough, and Dr. Yates, as well as by a corps of several 
clerical assistants. During November and December, 1922, 
a detailed report was made to the parents or guardian of 
each child of the gifted group, including definite information 
on grade of intellectual superiority, standing in each of the 
achievement tests in terms of "subject ages," information 
age, special abilities or weaknesses noted, comparison of the 
child's play interests with those of normal children of the 
same age and sex, appraisal of the child's reading, and 
general advice and comment on various points. A reduced 
copy of this blank follows. 



10 TRAITS OF GIFTED CHILDREN 

THE STANFORD UNIVERSITY GIFTED CHILDREN RESEARCH 

REPORT TO PARENTS: DATE , 

NAME OF CHILD - *>ATB OP BIRTH 

AGE WHEN TESTED GRADE WHEN TESTED ..DATE OF TEST . ... 

PARENT OR GUARDIAN ADDRESS 



(Parent should promptly notify Professor L. M. Terman, Stanford University, of 

any change of address,) 

I. GENERAL INTELLIGENCE. The child's grade of superiority is indicated by a cross. 

HGrd 1- Most rare (Abt. 5 or 10 children in 100,000 eql. or exceed this rank.) 

[1 Grd 2- Highly gifted A.. (Abt. 30 children in 100,000 eql. or exceed this rank,) 
[] Grd 3' Highiy gifted B.. (Abt. 90 children in 100,000 eql. or exceed this rank.) 
Q Grd" 4 ; Very superior A.. (Abt, 250 children in 100,000 eql. or exceed this rank.) 
f] Grd 5- Very superior B..(Abt. 500 children in 100,000 eql. or exceed this rank.) 

[] Grd 6* Superior A (Abt. 1,500 children in 100,000 eql. or exceed this rank.) 

[] Grd 7 Superior B (Abt. 7,500 children in 100,000 eql. or exceed this rank.) 

[] Grd.' 8: High average {Abt. 25,000 children in 100,000 eql. or exceed this rank.) 

Note. This rating has been given on the basis of standardized intelligence 
tests. It should be borne in mind, however, that intelligence tests are not infalli- 
ble; also, that success in school and in life is not determined entirely by one's 
intelligence. For these reasons the above rating is not to be taken as a sure indi- 
cation of what the child will later accomplish. It does suggest roughly what may 
be expected in case the child makes the most of his ability. 

Remarks ' 

H. SCHOOL ACHIEVEMENT. The child's achievement in the various sch<x*l subjects 
corresponds to the grades indicated below, 

Reading: corresponds to grade Nature and science: corresponds to grade,,.. ..... 

Arithmetic: corresponds to grade.....History ami civics: corresponds to grade,....,.... 

Language: corresponds to grade..,. Literary knowledge: corresponds to grmJe ..... 

Spelling: corresponds to grade ..All-round information; corresponds to grade,,,. 

Note. When school achievement is two or three gratles above that in which 
the child is located, an extra promotion is usually desirable, unless the child Is al- 
ready advanced in school very considerably beyond his years. When this Ss the 
case, or when achievement is only a little above the grade in which the child is 
located, it is often better to find extra work for the child to do than to give addi- 
tional promotions, This extra work may take the form of outside nwUttg or su- 
pervised study along any of the child's special lines of interest. Tenh*rs are 
usually glad to coo'perate in planning extra study, as it helps to keep the child busy 
and therefore more contented and interested in school work* Probably few chil- 
dren, however bright, should enter high school before the age of 11 or 12* or col- 
lege before the age of 15 or 16. Advancement much more rapid than this involves 
the risk of defective social development. 

Remarks , ,..,..,....... ,......,.,,., .... ......... . ...... ..< ,.., 

III. SPECIAL INTERESTS AND ABILITIES* (See next line for rating on evenness of J- 
terest and ability.) 

1. Markedly even, fairly even, somewhat uneven, markedly uneven, very 
or special. 

2. Special interest or abilities noted, if any..., , , ...... ,.,, ,.,..., 

3. Special weaknesses noted, if any ............................... .,...,... 

4. School subject preference - ............. ...,.,. ...... 

5. Occupational preference ......,.....,,,.,.,.. 

6. Remarks - .................I................ , 



HISTORY AND DESCRIPTION 11 

IV. PLAYS, GAKES AND AMUSEMENTS, 

1. Knowledge of, in comparison with that of average children of the same age: 

much greater, greater, equal to average, less, much less. 

2. The child's play interests appear to resemble those of children who are 

much older, somewhat older, same age, somewhat younger, much younger. 

3. The child's knowledge of plays, and the play preferences expressed, are in 

general markedly masculine, noticeably masculine, neutral, noticeably 
feminine, markedly feminine. 

4. The plays and games for which child has expressed preference indicate 

(a) That the play interests are noticeably non-social, socially normal, 
more than ordinarily social. 

(b) That the plays preferred are rather quiet or sedentary, ordinarily ac- 
tive, more active than average. 

5. Remarks , -.... 

V. READING. 

1. Amount of reading, in comparison with that of average children of the same 
age: very much greater, much greater, somewhat greater, equal to average, 
somewhat less, much less. 

2. Quality of books preferred: very superior, superior, average, rather inferior, 
very inferior. 

Note. If a bright child does not read much more than an average child, this 
does not necessarily mean that he ought to be urged to increase his reading. Some 
children show a natural preference for reading at a very early age; others prefer 
to spend their time on mechanical devices or in outdoor play* A taste for good 
reading should always be encouraged, but it is seldom advisable to urge the bright 
child to read when he strongly prefers to do other things. Most bright children 
read more than the average, and this is not objectionable if reasonably good books 
are selected and if it is not carried to such an excess as to rob the child of needed 
social contacts. 

Remarks - - 

VI. ADDITIONAL COMMENTS - 



VH. CONTINUATION OF THE STUDY. We are glad to report that the original grant of 
money to Stanford University from the Commonwealth Fund for conducting 
this investigation has been increased so as to make it possible to secure a thor- 
ough, standardised mt k dical examination of each child in our special group, also 
additional mental tests for more accurate diagnosis of special interests and 
abilities. We desire to carry out this extension of our work between Decem- 
ber* 1922, and July, 1923, and are hoping that parents will continue to give us 
the same splendid cooperation which has contributed so much to the success 
of our study thus far. It is hoped that this continuation of the study will help 
to bring us into closer personal touch with parents and children. PLEASE DO 

NOT FOHCBT TO NOTIFY US OF ANY CHANGE OF ADDRESS. 

VHL PUBLICITY. We have made and will continue to make every effort to avoid 
newspaper publicity concerning the results of this study. Several newspapers 
have tried to secure Information regarding some of our subjects, but we have 
always refused to give it Two or three unauthorized news items have been 
published, based mostly on the imagination of reporters. We sincerely hope 
that parents will cooperate with us in preventing newspaper exploitation of 
our subjects. The Information we have collected In the research is kept In 
locked files. 

[SIGNED].... 

Department of Psychology, 

Stanford University, California, 



12 TRAITS OF GIFTED CHILDREN 

Meanwhile, work under the second Commonwealth ap- 
propriation had already begun. This included the following 
divisions and personnel: 

1. Medical examinations, by Dr. Edith Bronson and Dr. 
Albert H. Moore, in cooperation with an advisory committee 
composed of Dr. Thomas Addis, Dr. Harold K. Faber, Dr. 
A. W. Hewlett, and Dr. W. E. Schaller, of the Stanford Uni- 
versity Medical School, and Dr. Ernest Gale Martin, Profes- 
sor of Physiology, Stanford University. 

2. Anthropometric measurements (37 in all), under the 
direction of Dr. Bird T. Baldwin, Director of Child Welfare 
Research, University of Iowa. 

3. Character and personality tests; methods prepared 
by A. S. Raubenheimer. 

4. Interest test; method prepared by Jennie Benson Wy- 
man. 

5. Preparation of a guide to children's reading, based on 
reading records of the gifted and control groups, by Mar- 
garet Lima. 

6. Study of the specialization of abilities, by James 0. 
DeVoss. 

7. A biographical and comparative study of the mental 
development of a representative group of 300 eminent in- 
dividuals, by Catharine M. Cox. 

The plan of the investigation called for the collection of 
about 65 pages of test and measurement data and about 35 
pages of questionnaire data, a total of approximately 100 
pages for each child. Practically all of this material was 
obtained for more than 90 per cent of the main experi- 
mental group of 643 subjects, and about half of it for nearly 
600 other gifted subjects. In addition, a large part of the 
material was also obtained for several hundred unselected 
children. For the intelligence, achievement, and anthropo- 
metric measurements norms were already available, but it 
was necessary to establish norms for the following: School 
Information Blank (including various kinds of educational 
data and teacher's ratings on twenty-five traits) ; the Inter- 
est Blank (filled out by the children) ; the two-months* read- 
ing record; the Information Test; the test of play interest 
and play knowledge; the Wyman Interest Test; and the 
tests of character and personality traits. This enormously 



HISTORY AND DESCRIPTION 13 

increased the labor involved in the investigation but pro- 
vided comparative data of the greatest value. It was not 
possible, unfortunately, to secure medical examinations or 
the data called for on the Home Information Blank for an 
unselected group. 

During the course of the investigation several different 
control groups were utilized, as it was not feasible to secure 
from a single group all the material that was desired. One 
group furnished norms for the School Information Blank, 
another for the Information Test, another for the interest, 
character, and personality tests, another for the reading rec- 
ord, and another for the Interest Blank and test of play in- 
terest and play knowledge. Each of these groups numbered 
from 600 to 800 children. 

After the results of the second year's work had been 
summarized, another report was issued to parents of the 
gifted group. The form used in this report follows. A third 
report, that summarizing the most important data from the 
physical measurements, is reproduced in Chapter VII. 

THE STANFORD UNIVERSITY GIFTED CHILDREN RESEARCH 
REPORT TO PARENTS ON INTEREST AND PERSONALITY TESTS 

Bate of this report , 

Name of child...... Date of birth,...., 

Parent or guardian Date of test 

Address .....,...,, , ... 

(PLEASE NOTIFY PROFESSOR TRRMAN PROMPTLY OF ANY CHANGE OF ADDRESS.) 

Note. These tests have only recently been standardized, and sufficient time 
has not yet elapsed to permit a thorough checking with other findings. In view of 
this fact, the results herein reported must he considered somewhat less authori- 
tative than the results of intelligence tests or of tests of school achievement. How- 
ever, they have been found to have a fair degree of dependability in the majority 
of cases. 

In each test the child's rating has been compared with that of a group of un- 
selected children of the same age and sx In case of an unfavorable report on any 
characteristic, the parent should observe the child carefully. If the report seems 
to be justified, it may be advisable to undertake special training for the purpose 
of bringing about improvement. 

I. CHILD'S TENDENCY TO UNDERESTIMATE OH OVERESTIMATE HIS OWN KNOWLEDGE 
AND ACHIEVEMENT. (Bating is based upon a comparison between the child's state- 
ments as to his knowledge of certain facts, books which he has read, etc., and his 
score on lists of questions which have been so devised as to show his actual 
knowledge about these same facts.) 

The child's ratings on this test indicate that: 

H* shows a marked tendency to overstate ; 

Overstates, but not more than does the average child of his age and sex ; 

Estimates his knowledge accurately; 

Has a tendency to underestimate his own knowledge. 



14 TRAITS OF GIFTED CHILDREN 

II SOCIAL ATTITUDES. (Based upon child's preference as indicated by Us choice 
of books from a list of descriptive titles, his choice of companions from a list of 
described characters, and his expressed attitude toward various social organi- 
zations.) 

The ratings on these tests indicate a social attitude which is: In general very 
desirable, similar to average for age and sex, less desirable than average. 

Remarks; ' '"""" 



III. RELIABILITY OF PERFORMANCE IN THE FACE OK TEMPTATION NOT TO FOLLOW TH* 

RULES LAID DOWN : 

More than average, equal to average, less than average. 

Remarks: - " ' ' " 

IV. EMOTIONAL STABILITY. (Based on answers to 85 questions designed to 
bring out eccentricities of emotional attitude.) 

As compared with average children of the same age and sex, the child seem* 
to be: well balanced, of average stability, of less than average stability. 



Remarks * 



V. INTERESTS. (Based on a word association test.) 

Note. Children as well as adults differ greatly in the general type of their in- 
tests. Some are predominantly active, preferring always the role of the doer to 
that of the onlooker, others show especially strong social interests, still others re 
interested primarily in intellectual matters. There are some whose interests are 
not especially strong in any of these lines, and others whose interests *tre strongly 
developed in all of them. The results of the tests arc as given below. 

1. INTELLECTUAL INTERESTS. This child's interests are : 
Decidedly intellectual 

Somewhat more intellectual than average 

Equal to average in intellectuality 

Somewhat less intellectual than average 

Decidedly non-intellectual 

Note. This rating has to do with NATURE OP INTERESTS only, not with mental 
ability. A child whose general mental ability is very high may nevertheless s&ow 
INTERESTS which are more pronounced along social or active lines than along in- 
tellectual lines. 

2. SOCIAL INTERESTS. This child's interests are: 
Decidedly social 

Somewhat more social than average 
Equal to average in sociability 
Rather non-social 
Decidedly non-social 

3. ACTIVITY INTERESTS, This child's intere&ts are: 
Decidedly strong along activity lints 

Rather stronger than average 
Equal to average in activity 
Rather less than average 
Decidedly less than average 



Remarks : 



[SIQNRD] , ,.... , ,,....,.... 

Department of Psychology, 

Stanford University, California. 



HISTORY AND DESCRIPTION 15 

Miss Goodenough served as chief research assistant 
throughout the second year, assisted in the standardization 
of the character, personality, and interest tests which had 
been devised and validated by Mr. Raubenheimer and Mrs. 
Wyman, directed the application of these tests in the spring 
of 1923 to the gifted group, and from June, 1923, to Septem- 
ber, 1924, assisted in the preparation of the report. In this 
work she was assisted a part of the time by Miss Helen 
Marshall and Miss Alta Williams. 

The choice of assistants for the second year's work 
proved to be no less fortunate than for the first year. Both 
by training and experience Dr. Bronson and Dr. Moore were 
ideally fitted for the medical examining. Dr. Baldwin had 
long been engaged in studies of the physical growth of chil- 
dren from birth to maturity. Miss Cox, Mr. DeVoss, Mr. 
Raubenheimer, Mrs. Wyman, and Miss Lima were advanced 
graduate students in psychology and education at Stanford 
University, and, with the exception of Miss Lima, were 
working on Ph.D. dissertations in connection with this re- 
search. 

Throughout its course the investigation owes much to 
the devoted work of the field assistants, who constantly 
worked overtime, industriously ran down every clue that 
promised to yield new subjects or useful information, and 
by their tactful dealings with parents and teachers secured 
from home and school the cooperation without which the 
study could not have proceeded. Of the 649 children who 
qualified for our main group, the parents of only 6 refused 
cooperation. The total time-cost of the assistance which 
parents have rendered has been very great. To fill out the 
schedule of information called for in the Home Blank is 
alone a task requiring several hours. In addition, approxi- 
mately four half-days were required to take the child to 
the educational tests, the personality and interest tests, the 
medical examination, and the anthropometric measure- 
ments. Almost invariably the parents have shown an in- 
terest in the outcome of the study apart from any help they 
might hope to receive from it. 

The purpose and methods of the various divisions of the 
research are set forth in their respective sections of this re- 
port and need not be described here. It may be stated, how- 
ever, that each part of the study was carried out essentially 



16 TRAITS OF GIFTED CHILDREN 

as planned, and with results which are believed in each case 
to throw considerable light on the problems under investi- 
gation. 

This volume deals with only the first stages of an inves- 
tigation which will be continued, with the same subjects, 
for many years. The material that has been gathered is far 
too extensive to be summarized in satisfactory detail in a 
single volume. A brief clinical description of each case 
would alone have required a volume as large as the present 
one. Every section of the data would have warranted more 
thorough and detailed treatment than it has thus far been 
possible to give. The data already collected can be made 
the basis of numerous minor studies, and the follow-up 
work which will be carried on indefinitely will increase 
many times the value of the original material. The present 
purpose is to show in what traits, and to what extent, a rep- 
resentative group of intellectually superior children differs 
from a group of unselected normal children. The task of 
the future will be the comparison of promise and perform- 
ance. In the fulfillment of this task, new light will be thrown 
upon the prognostic significance of the test scores and of 
the other records which have been secured. When our cases 
have thus been read backward, so to speak, it will be easier 
to read other cases forward; to predict and prescribe in the 
light of long-range knowledge. Another unlimited field for 
future research is in the genealogical study of the families 
represented in our gifted group. It is to be hoped that finan- 
cial support will be found for this and for other studies 
which it has thus far been impossible to undertake. 

But increased knowledge of the origin and of the phys- 
ical and mental traits of gifted children is not an end itself. 
When the sources of our intellectual talent have been de- 
termined, it is conceivable that means may be found which 
would increase the supply. When the physical, mental, and 
character traits of gifted children are better understood it 
will be possible to set about their education with better hope 
of success. Educational experiments in this line are already 
being undertaken in the public schools in ever increasing 
number, but these need to be supplemented by privately 
endowed undertakings which permit of greater freedom for 
experimentation. Surely, in a nation of a hundred million 
people there should be one or more schools of this kind; 



HISTORY AND DESCRIPTION 17 

schools which would be unhampered in the selection of such 
pupils as the experiment called for, which would be free 
to follow without hindrance the lead of experimental evi- 
dence, and financially in position to allow for long-range 
planning. In the gifted child. Nature has moved far back 
the usual limits of educability, but the realms thus thrown 
open to the educator are still terra incognita. It is time to 
move forward, explore, and consolidate. 



CHAPTER II 

METHOD AND RESULTS OF THE SEARCH 
FOR SUBJECTS 

THE PROBLEM 

The purpose of the search was to locate subjects of a 
degree of brightness that would rate them well within the 
top one per cent of the school population. It was also de- 
sired to secure subjects who would be as fairly as possible 
representative of all gifted children of the degree of bright- 
ness set as a standard for selection. It is obvious that con- 
clusions regarding the characteristics of gifted children in 
general wall be valid only to the extent to which the latter 
requirement has been met. Too often physicians and psy- 
chologists, as well as laymen, have based their conclusions 
regarding the supposed abnormality of intellectual pre- 
cocity on selected cases. 

It was not considered feasible to attempt to locate the 
1,000 brightest children in California, desirable as that would 
have been. Apart from the difficulties involved in equating 
the intellectual superiority of bright children of different 
ages, such a plan was entirely impracticable because of the 
size of the state (158,000 square miles) and the large school 
population (approximately 500,000 in grades 1 to 8). The 
cost of intelligence test blanks for a half million children 
would alone have greatly exceeded the funds available for 
the entire research. The labor of giving and scoring the tests 
would probably have brought the cost of covering the entire 
state by this method to many times the amount of money 
available for the study. The limitation of the survey to the 
larger cities was therefore a matter of necessity. Traveling 
expenses were thus reduced to a minimum, as well as the 
time required for making the necessary arrangements with 
school officials. Such limitation, however, has undoubtedly 
affected the findings in various important ways, especially 
with respect to racial and social origin of the subjects, 
their scores on the various achievement tests, their grade 

19 



20 TRAITS OF GIFTED CHILDREN 

advancement, their interests, their reading, and their recre- 
ational habits. 

The next problem was to secure a group of subjects who 
would be as representative as possible of all gifted children 
in the territpry covered. A satisfactory solution of this prob- 
lem would have required the application of a perfect mea- 
sure of intelligence to all the children, A perfect measure 
was not available, and even if it had been, the cost of its 
application would have been too great. It was necessary, 
however, to find some kind of criterion for the selection of 
an experimental group. The leading possibilities considered 
were teachers* ratings, age-grade status, achievement tests, 
and intelligence tests. All of these criteria have been used 
from time to time, singly or in combination, by various in- 
vestigators. One would hardly expect any two of them to 
yield the same subjects out of a given school population* 
and it is possible that the subjects selected by any one of the 
methods would differ appreciably in general characteristics 
from those selected by any other. The rank order of the 
above criteria for validity *s probably as follows: (1) intel- 
ligence tests, (2) achievement tests, (3) age-grade status, 
(4) teachers' ratings. 

The faults of subjective ratings have been sufficiently 
exposed in numerous investigations to show their great un- 
reliability when used alone. Such ratings are usually based 
too largely on the child's class work and are almost certain 
to weigh too lightly the age factor. Age-grade status is per- 
haps a better criterion, but its value is limited (1) by the 
variable standards in different cities, or even in different 
schools of the same city, and (2) by the difficulty of equat- 
ing various degrees of acceleration for children of the dif- 
ferent ages. The first-named objection is the most serious* 
Moreover, the age-grade status of a given child rests ulti* 
mately upon a teacher's rating; that is, upon a judgment as 
to fitness for promotion. It is also likely to be affected by 
age of entering school, regularity of attendance, adaptabil* 
ity to school requirements, and various other factors having 
nothing to do with intelligence. Achievement tests, although 
they are an objective method, were not seriously considered, 
as they would have been more costly than intelligence tests 
and inferior to them as measures of native ability. 



SEARCH FOR SUBJECTS 21 

SELECTION OF THE MAIN GROUP 

It was decided to use intelligence tests as the final cri- 
terion for inclusion of subjects in the main experimental 
group. However, since it was not feasible to test the entire 
school population, even by an abbreviated test, it was nec- 
essary to use a preliminary sifting method to determine 
what children should be tested. The method adopted em- 
ployed both teachers' ratings and age-grade status. In grades 
3 to 8 the procedure involved three steps, as follows : 

First step. From each class (composed ordinarily of 30 
to 50 pupils) from one to five children (usually four) were 
selected for a mental test by the plan shown in the follow- 
ing blank (printed here in reduced type) . 

BLANK FOR THE SELECTION OF GIFTED CHILDREN 
REGULAR CLASS TEACHER 

TO BE FILLED OUT AND RETURNED TO THE PRINCIPAL'S OFFICE BEFORE 

Name of teacher Grade taught 

School City 

To THE REOULAH CLASS TEACHER: Below, write the names of from one to three 
pupils whom you regard as the most INTELLIGENT in your class or classes. If your 
class as a whole is distinctly superior to the average, three names may be given. 
If your class as a whole is distinctly inferior, give one name only. If your class 
is about average, give two names. 

Do not base your judgment of intelligence upon school marks alone. Im- 
portant qualities to consider are quickness and accuracy of mental grasp, origi- 
nality, ability to reason clearly about new and difficult problems, breadth and 
accuracy of information, intellectual curiosity, command of language, common 
sense, and independence of judgment. 

Take age into account. Of two pupils who seem to be about equally excep- 
tional, but who differ one or more years in age, the younger is probably the more 
intelligent. 

Do not underrate the bright child who is shy, or lacks industry, or stands low 
in deportment * 

It is permissible to name one child of moderate general ability, provided such 
child is very exceptionally gifted in some special line, as in music, drawing, 
modeling, mechanical ingenuity, science, mathematics, composition, etc. 

I. Most intelligent: Name, .Age (yrs. and mos.) Grades... 

Selected because of all-round intelligence or because of some particular talent? 

(Underline) 

If because of particular talent, In what line or lines? 

Is this child*s ability "very extraordinary," "decidedly superior," or only 

"slightly superior"? 

2* Next most intelligent; Name Age (yrs. and mos.) Grade..... 

Selected because of all-round intelligence or because of special talent? (Under- 
line) 

If because of some special talent. In what line or lines? 

I No. 2 almost as exceptional as No. 1, or considerably less exceptional? Un- 
derline) 



22 TRAITS OF GIFTED CHILDBEN 

3. Third most intelligent: Name Age tyrs. and mos.) ...Grade 

(Third name to be given only in exceptionally superior classes.) 

Selected because of all-round intelligence or because of special talent? tUnder- 

If because of some special talent, in what line or lines? - . ... 

Is No. 3 almost as exceptional as No. 2, or considerably loss exceptional? lUn- 
derline) 

4. Below name the YOUNGEST pupil in each grade or half-grade represented in your 

room: 

Grade or half-grade. Youngest pupil Age tyrs. and mo*.).... 

Grade or half-grade Youngest pupil... . Age tyrs. and mos.)... 

Grade or half-grade, Youngest pupil Age (yrs. and mos.^.. 

Grade or half-grade .Youngest pupil Age tyrs, and mos.) 

5. In case you taught in this building during the half-year preceding this, give the 

name of the brightest child in your class THEN... What grade is 

that child in now? 

On the above blank each regular classroom teacher made 
her nominations. In general, about 6 or 8 per cent of the 
pupils in grades 3 to 8 were tested, but the proportion va- 
ried from school to school. In a few of the best schools as 
high as 20 per cent of the pupils enrolled were tested; in the 
poorest schools, as low as 2 per cent. 

Second step. Next, the nominees from several class- 
rooms, often those from an entire school building, were as- 
sembled in a group and given the National Intelligence Test, 
Scale B, Form I. Those who ranked in the top 5 per cent of 
unselected children of their respective ages were retained 
for further study. This standard proved to be too high, and 
was later lowered to the ninetieth percentile, i, e*, top 10 
per cent, in the case of children who, judged by teachers* 
ratings or age-grade status, seemed to be promising cases. 
Occasionally the standard was lowered to the eighty-fifth or 
even eightieth percentile, especially in the case of children 
of ages 7 and 8. A 7 year old was found, of Binet IQ above 
140, who rated only at the eighty-fifth percentile on Na- 
tional B. By our failure to give a Binet test to all above the 
eightieth percentile on the National, a few cases were un- 
doubtedly lost who would have earned an IQ of 140, 

The National age norms used in making this segrega- 
tion were those found by the writer for the Vallejo, Cali- 
fornia, public school children, 1 These were practically the 
only norms available which were based upon tests of an 

^ee Journal of Educational Research, 1921, pp. 124-132. 



SEARCH FOR SUBJECTS 



23 



entire school population of ages eight to fifteen inclusive, 
and were used in preference to the age norms published in 
the National Intelligence Test Manual. The norms used were 
as follows: 



Percentile Scores by Age 
10 11 12 13 14 



15 



10% 
25% 
20% 
25% 



equal or exceed 



85 100 114 129 142 154 161 165 

72 93 104 123 137 149 157 159 

63 87 100 118 132 144 153 156 

57 83 97 114 126 138 150 153 

51 78 93 109 122 134 147 150 



Third step. Pupils retained by Step 2 were next given 
an abbreviated Stanford-Binet test. Two abbreviations were 
used, one for children of foreign parentage, the other for 
children of non-foreign parents. They were as follows: 



Foreign 

1. Compares lines. 

2. Discriminates forms. 

3. Counts 4 pennies. 

4. Copies square. 



1. Comparison of weights. 

2. Colors. 

3. Aesthetic comparison. 
5. Patience. 



1. Right and left 

2. Mutilated pictures, 

3. Counts 13 pennies. 
5. Coins. 



L Fingers. 

2. Picture description, 

4. Bow knot. 

G. Diamond. 



1. Ball and field 

2. Counts 20 to 1. 

3. Comprehension. 

4. Similarities. 



Non-Foreign 



YEAR IV 



1. Compares lines. 
3, Counts 4 pennies. 
5. Comprehension. 



YEAR V 

1. Compares weights. 

2. Colors. 

4. Definitions. 

YEAR VI 

1. Right and left. 

2. Mutilated pictures. 
4. Comprehension. 



YEAH VII 
1. 



4. 
6. 



Fingers. 
Bow knot. 
Diamond. 



YEAR VIII 

2. Counts 20 to 1, 

3. Comprehension. 
6. Vocabulary. 



24 TRAITS OF GIFTED CHILDREN 

Foreign Non-Foreign 

YEAR IX 

1 Date. 3. Change. 

2* Weights 5* Three words. 

3! Change." Alt. No. 1. Naming months. 

4. Four digits reversed. 

YEAR X 

2. Absurdities. 1- Vocabulary. 

3. Designs. 3. Designs. 

5. Comprehension. 5. Comprehension. 

6. Sixty words. 

YEAR XII 

3. Ball and field. 1. Vocabulary. 

6. Five digits reversed. 4. Dissected sentence. 

7 Picture interpretation. 5, Fables. 

8. Similarities. 7. Picture interpretation. 

YEAR XIV 

2. Induction. 1- Vocabulary. 

3. President and king. 2. Problems of fact. 

5 Arithmetical reasoning. 5, Arithmetical reasoning. 

6! Clock. 6. Clock. 

YEAR XVI 
2. Fables. 1- Vocabulary. 

4. Boxes. 2. Fables. 

5. Six digits reversed. 5. Six digits reversed. 

6. Code. 6. Code. 

YEAR XVIII 

2. Paper cutting. 1. Vocabulary. 

3. Eight digits, direct order. 4, Passages. 

5. Seven digits reversed. 5. Seven digits reversed. 

6, Ingenuity. 6. Ingenuity, 

The selection of tests for the abbreviated Stanford-Binet 
was made with considerable care and took account of diag- 
nostic value, variety, brevity, and ease of administration. 
Judgment regarding diagnostic value was based chiefly on 
correlations of each test with mental age on the complete 
scale. Such correlations, in terms of per cent of pupils of 
each mental age passing each test, were available in two 
Stanford studies. 1 The use of this method is justifiable in 

*(a) For school children, Tennan et al: The Stanford Revision and 
Extension of the Binet-Simon Scale for Measuring Intelligence, Warwick 
and York, p. 164 ff. 

(b) For 400 adults, in an unpublished master's tbesis by H. E, Kaoi- 
lin, Stanford University. 



SEARCH FOR SUBJECTS 25 

view of the fact that the aim is to get an abbreviation which 
will yield approximately the same results as the entire scale. 

Non-dependence upon language was, of course, the most 
important consideration in the abbreviation for children of 
foreign parentage. In the other abbreviation, also, care was 
taken to avoid the too exclusive use of verbal or "schoolish" 
tests. It was thought that the method of nomination would 
probably over-weight the language factor, and that the in- 
telligence tests ought to err in the opposite direction, if at 
all. For this reason, in Step 2, National Scale B was used 
almost exclusively, in preference to Scale A, the latter being 
more verbal. It is probable that the large use made of non- 
verbal tests gave a group slightly different from that which 
would have resulted from the use of tests depending more 
upon the language factor, perhaps on the whole group 
slightly less gifted intellectually. On the other hand, the 
method gave the much-desired assurance that the experi- 
mental group would not be characterized primarily by spe- 
cial ability in language. 1 

If the vocabulary test seems over-weighted in the abbre- 
viations used, it is only necessary to point to the correlation 
of .85 to .95 which have been found between vocabulary 
score and mental age on the entire scale. 2 The amount of 
this correlation was deemed to justify rejection of some 
candidates on vocabulary test alone. This was frequently 
done in the case of non-foreign children of seven years or 
older, though never with children below seven. Usually, 
the vocabulary test was given first, and in case time was 
limited the examiners were permitted to reject without fur- 
ther test those having a vocabulary quotient of less than 
125. After some experience with this rule the standard 
usually followed was a vocabulary quotient of 120. Exami- 
nation of the table on page 308 of Terman's The Intelli- 
gence of School Children will show that the 120 rule is 
reasonably safe. The following age norms were used in 
computing vocabulary quotients : 

Mental Age 789 10 11 12 13 14 15 16 17 18 19 

Median Vocabulary 13 18 23 30 35 41 46 51 57 62 67 73 75 

^National Scale B consists of the following five tests: arithmetical 
computation; general information; sentence meaning; analogies; compari- 
son of numbers,, forms, and words. 

*Terman, Lewis M,: The Intelligence of School Children, p. 308. 



26 TRAITS OF GIFTED CHILDREN 

Pupils attaining an IQ of 130 or more on the abbreviated 
scale were given a complete examination. The correlation 
between the entire scale and the abbreviated scale is about 
.95 for unselected children of a given age, which means that 
the 130 rule is fairly safe. However, very few cases were 
rejected on an abbreviated test whose IQ was as high as 
125. Older children were ordinarily not rejected on an ab- 
breviated test if the IQ was as high as 120. It was possible 
to economize time in giving the abbreviated scale by begin- 
ning with vocabulary, going next to memory for digits, then 
working backward with other tests, beginning with the dif- 
ficult and proceeding to the easy until enough tests had been 
missed to disqualify the subject, at which point the exami- 
nation was ordinarily abandoned. 

IQ 140 on the complete scale was set as the provisional 
lower limit for inclusion in the case of children under eleven 
years. For older children an allowance had to be made for 
the fact that the brightest children of eleven years or older 
are graded too low^by the Stanford-Binet The standards 
set were as follows : 

Age Score 

Below 11 years IQ 140 

11 to 11% " " 139 
11% to 12 " " 138 

12 to 12% 4t fc " 137 
12% to 13 ** " 136 

13 to 13% " " 13-1 
13% to 14 " fc * 132 

Binet tests were not given to pupils above the age of 14. 
The Terman Group Test was to be used instead, but it 
turned out that few fourteen year old children of very 
superior intelligence were found in the grades below the 
high school. 

Such was the method of selection in grades 3 to 8* The 
method in grades 1 and 2 was at first exactly the same, ex- 
cept for the omission of the National test. During the first 
two months of the survey all nominees in grades 1 and 2 
were given an abbreviated Binet. During this period ap- 
proximately three-fourths of the time of the field assistants 
was devoted to the first two grades. Moreover* the results 



SEARCH FOR SUBJECTS 27 

here were on the whole less satisfactory than they were in 
the upper grades, because the teachers' nominations were 
more often in error. Accordingly, after December 1, 1921, 
grades 1 and 2 were no longer canvassed by the use of nomi- 
nation blanks. Instead, the field assistant thereafter merely 
visited each primary teacher and enquired whether she had 
any pupils of very outstanding ability or exceptionally 
under age for the grade, and such pupils only were tested. 

A good many cases were discovered by other than the 
usual method. Whenever possible, sibs of cases already lo- 
cated were tested. This netted a considerable number and 
accounts for the majority of those in the main group who 
were below school age. 1 

A few pre-school cases were located as a result of casual 
information. Sheer accident accounted for perhaps a half 
dozen school cases. Some of these accidents were rather 
surprising. In one case the teacher in nominating the young- 
est child in her room reported by accident the child whose 
name was adjacent to that of the youngest child on the roll. 
This proved to be the only child in the 300 pupils of that 
building who tested as high as 140 IQ! Another child who 
met the standard was brought to the test by a child messen- 
ger. It was afterward learned that the messenger had by 
some mistake brought another child than the one the teacher 
had intended to send. Another subject who qualified was 
the second-youngest in his class and was only sent to the 
test because the youngest pupil was absent. 

Loss OF SUBJECTS 

Accidental discoveries of the kind mentioned above were 
frequent enough to suggest that a considerable proportion 
of gifted pupils were being missed. A test experiment was 
therefore arranged through the cooperation of Dr. James L. 
Stockton, of the State Teachers' College, Santa Barbara. Dr. 
Stockton had 33 teachers in grades 2 to 6 of seven schools 
in the city of Santa Barbara make their nominations in regu- 
lar manner. He then gave National B, not only to the pupils 

'Additional sibs to the number of 48 were tested during 1922-23, at 
the time the character and personality tests were given, and of this num- 
ber, 19 qualified. These have not been added to the 643 cases of the main 
group In the statistical treatment of most of the data in this report. 



28 TRAITS OF GIFTED CHILDREN 

nominated, but to the entire school population of those 
grades. Those reaching the ninety-fifth percentile on the 
National were all to be given a Stanford-Binet as in our 
main study. In about half the cases, however, this was not 
possible. The results showed that of the eight pupils qualify- 
ing on the data collected, three Jiad not been named on the 
nomination blanks. Of the five who qualified, four were first 
choices (one of whom was also youngest), and one was sec- 
ond choice and also youngest There were five others, not 
given a Binet test, who would almost certainly have qualified 
had the data been complete, as all of them reached the 
ninety-fifth percentile score on National B. One of these 
had been nominated as brightest and youngest, one as sec- 
ond brightest, one as second brightest and youngest, and 
two as youngest only. Assuming, conservatively, that only 
four of these five most promising incomplete cases would 
have qualified, the total number would have been twelve 
cases, of whom three, or 25 per cent, would have been 
missed by the method of search that gave us our 643 regu- 
lar cases* 

If the Santa Barbara data could be taken as typical it 
might be inferred that our main survey resulted in the lo- 
cation of only about 75 per cent of the subjects who could 
have met the standards set if all had been tested. The field 
assistants estimate, however, that the efficiency was nearer 
90 per cent. As an additional check, one school of 350 pupils 
in Los Angeles, and another in San Francisco of 800 pupils, 
were re-sifted. These two schools were among the very best 
in their respective cities. In the first sifting the Los Angeles 
school had netted 12 cases (1 for every 25 enrolled), and the 
San Francisco school 28 cases (1 for every 28 enrolled). 
In each school several teachers, after the first sifting, 
protested that there were other children in the school who 
were bright enough to qualify. Accordingly, the teachers 
were all given a second set of nomination blanks and 
requested to nominate the youngest, the brightest, and the 
second brightest of those who had not been nominated in 
the first sifting, Tests were then made in the usual manner. 
In the Los Angeles school not a single new case qualified, 
but in the San Francisco school ten reached the standard. 
Of the total 50 cases thus located in these schools, 20 per 



SEARCH FOR SUBJECTS 29 

cent would have been missed but for the second survey. 
It is entirely improbable, however, that the general loss was 
anything like as great, for the chances of loss would be lower 
in schools attended by average or inferior population. 

After a little experience the field assistants adapted the 
method of search somewhat according to the type of school 
in which they were working, and as a result were both able 
to save time and to make the search more effective. In the 
best schools more pupils were tested than the scheme called 
for, while in the poorest schools it was not necessary to 
test so many. In the good schools much testing was done in 
grades 1 and 2, but if a large school had netted no cases in 
grades 3 to 8 it was deemed safe to omit grades 1 and 2. 

It is believed that a considerable loss was incurred by 
our inability to include private schools in the survey. Judg- 
ing from the number and character of the private schools 
in the three centers, this loss was probably greatest in Los 
Angeles, where such schools are numerous and patronized 
by the superior social classes, and least in San Francisco, 
where they are chiefly parochial. 

SCHOOL POPULATION CANVASSED 

The main search in the cities of Los Angeles, San Fran- 
cisco, and the East Bay cities of Oakland, Berkeley, and Ala- 
meda, yielded 643 subjects, not including 6 whose parents 
refused to cooperate. These 643 ca$es were distributed as 
follows: Los Angeles, 285; San Francisco, 176; East Bay, 
182 (81 of these from Berkeley and Alameda, 101 from Oak- 
land). The proportion of cases found to school population 
canvassed in each section was roughly as follows; Los An- 
geles, 1 to 330; San Francisco, 1 to 235; Oakland, 1 to 225; 
Berkeley and Alameda, 1 to 100. 

Figures in Table 1 show approximate numbers actually 
covered in the main survey; they do not include schools or 
classes not canvassed. Only about one-half the population 
of grades 1 and 2 was covered. Probably 8,000 of the 12,000 
school population of Berkeley and Alameda (grades 1 to 8) 
were also canvassed. Since 649 regular cases qualified from 
about 168,000 covered in the main survey, the proportion 
is about 1 for each 258. If 20 per cent of the cases who could 
have qualified were missed, the number qualifying should 



30 TRAITS OF GIFTED CHILDREN 

have been about 812, which would have given a proportion 
of 1 for each 200, or about one-half of one per cent. 

Accurate figures are not available for the school popu- 
lation covered in other ways than by the main survey; that 

TABLE \ 
SCHOOL POPULATION CANVASSED 

Grades Los Angeles San Francisco Oakland Total 



1 


11,582 


3,470 


1,922 


16,974 


2 


7,744 


3,011 


1,705 


12,460 


3 


14,878 


6,278 


3,541 


24,697 


4 


12,808 


6,247 


3,453 


22,508 


5 


12,118 


5,592 


3,131 


21,241 


6 


11,206 


5,645 


3,052 


19,903 


7 


10,600 


5,518 


3,088 


19,206 


8 


14,377 


5,053 


2,922 


22,352 


9 




471 




471 



Total 95,313 41,685 22,814 159,812 

Total enrollment "] 

in grades 1 to 8 

in the schools I 104,330 43,100 26,844 174,274 

canvassed 

(1921-1922) I 

is, in the high schools throughout the State and in various 
city schools covered by volunteer assistants. Probably 100,000 
is a conservative estimate, which, added to the 168,000 cov- 
ered in the main search, would give a total school population 
of more than a quarter of a million. 

ANALYSIS OF THE SOURCES OF INFORMATION WHICH 
YIELDED THE MAIN EXPERIMENTAL Gaoup 

Interesting facts were brought out by a comparison of the 
number of subjects in the main experimental group who 
were located as a result of various kinds of information, in- 
cluding nomination blanks, sib relationship to a child al- 
ready located, previous tests, special recommendation, etc,; 
also, in the case of those discovered by the use of the nomi- 
nation blank, the relative proportions recommended as 
"most intelligent" in the class, "next most intelligent," "third 
most intelligent," "youngest," "brightest in preceding class," 
etc. The following data are for 644 cases of the 649 located 
in the main survey. 



SEARCH FOR SUBJECTS 31 

Per cent 

MEANS OF DISCOVERY N f g "oup 

1. By the use of nomination blanks 447 69.4% 

2. High record in an earlier test 83 12.9% 

3. Special recommendation 48 7.5% 

4. Sib relationship to child who had already qualified 41 6.4% 

5. By accident or unknown means 15 2.3% 

6. By repeated survey of two schools 10 1.5% 



644 100.0% 

The fact that about 30 per cent of the entire group were 
located by other means than through the use of nomina- 
tion blanks is due partly to the fact that after the first. two 
months the use of such blanks below the third grade was 
abandoned and that thereafter children in grades 1 and 2 
were only tested as a result of special recommendation. A 
good many other cases were discovered because of sib re- 
lationship to a child who had qualified. 

The following data show the relative number who re- 
ceived various kinds of nominations. In this case the per 
cents are based upon the 447 children who were recom- 
mended in nomination blanks, and in each case give the per 
cent of this group who received a particular kind of recom- 
mendation and no other. It does not include those who were 
nominated in two ways, e.g., as "most intelligent" and also 
as "youngest." 

Per cent 

KIND OF NOMINATION N nominated 

1. "Most intelligent" in class 70 15.7% 

2. "Next most intelligent" in class 42 X 9.4% 

3. "Third most intelligent" in class 18 4.0$. 

4. Youngest in class 88 -(7% 

5. From teacher of previous half-year 17 3.8% 

The remainder of the group are accounted for as follows: 

N Per cent 

Two kinds of nomination 157 35.1% 

Three kinds of nomination 55 ^?,^ 

** ~ j ti 

The above figures show that of those discovered as -a 
result of a single kind of nomination, the largest number 
belong to the "youngest" group and the next largest num- 



32 TRAITS OF GIFTED CHILDREN 

her to the "most intelligent" group. However, more than 
a third of all who were nominated received nominations of 
two kinds. 

An analysis has also been made of each major group, as 
shown below. 

Per cent of 

nominated 

Group nominated as brightest: N group 

1. Total nominated as brightest 230 51.5% 

2. Nominated as brightest, no other nomination 70 15.7% 

3. Nominated as brightest, also as youngest 105 23.5% 

4. Brightest, also nominated by previous teacher 103 23.0% 

5. Brightest, also one other kind of nomination 112 25.1% 

6. Brightest, also two other kinds of nomination 48 10.7% 

Group nominated as second brightest: 

1. Total nominated as second brightest 83 18.6% 

2. Second brightest, no other nomination 42 9.4% 

3. Second brightest, also youngest 29 6.5% 

4. Second brightest, also nominated by previous 

teacher 19 5.3% 

5. Second brightest, and one other nomination 34 7.6% 

6. Second brightest, and two other nominations 7 1.6% 

Group nominated as third brightest: 

1. Total nominated as third brightest 22 4.9% 

2. Third brightest, no other nomination 18 4.0% 

3. Third brightest, also youngest 3 0.7% 

4. Third brightest, also nominated by previous 

teacher 1 0.2% 

5. Third brightest, and one other nomination 34 7.6% 

6. Third brightest, and two other nominations 0,0% 

Group nominated as youngest: 

1. Total nominated as youngest 232 51.9% 

2. Youngest, no other nomination 88 19.7% 

3. Youngest, and one other nomination 89 19.9% 

4. Youngest, and two other nominations 55 12.3% 

Group nominated by previous teacher: 

1. Total nominated by previous teacher 147 32.9% 

2. Nominated in this way and no other 17 3.8% 

3. This, and one other nomination 75 16.8% 

4. This, and two other nominations 55 12.3% 

Perhaps the most important single finding is that nomi- 
nation as youngest yielded more subjects who would other- 
wise have been missed than any other kind of nomination, 



SEARCH FOR SUBJECTS 33 

19.7 per cent of the total nominated group. Nomination as 
brightest yielded 15.7 per cent who were nominated in no 
other way. In other words, if one would identify the bright- 
est child in a class of 30 to 50 pupils it is better to consult the 
birth records in the class register than to ask the teacher's 
opinion. This finding has a very high reliability, as it is 
based on the nominations made by approximately 6,000 
teachers. 

The fact that nomination as second brightest yielded 42 
subjects (9.4 per cent of entire nominated group) who 
would not otherwise have been discovered, and that nomi- 
nation as third brightest yielded 18 subjects, or 4.0 per cent 
of entire nominated group who qualified, strongly suggests 
that a similar number may have been missed by failure to 
call for second youngest and third youngest. If so, our group 
of 643 should have been larger by about 60, in which case 
the total would have been a little over 700. This would mean 
a loss of about 8.6 per cent. It is probable that the loss from 
this failure was between 7 and 10 per cent of the total num- 
ber of availables. 

However, to have included the second youngest and third 
youngest would have necessitated giving a National test to 
12 or 15 per cent of all the pupils. Possibly it would have 
been still better to have given a five-minute opposites test 
to all the pupils as a means of preliminary sifting. This 
method was considered and was rejected only for the rea- 
son that its use would have given rise to the criticism that 
the gifted group so obtained would tend to belong to the 
verbal type. 

One may conclude that the method of selection em- 
ployed, although far from ideal, probably led to the dis- 
covery of at least 80 per cent and possibly 90 per cent of all 
the cases who could have qualified in the school popula- 
tion canvassed. One can only surmise how the undiscovered 
cases would have differed from our experimental group in 
sex proportion, mean intelligence, social origin, personal 
characteristics, etc. They would almost certainly have been 
found a little less accelerated in school. Some would be ex- 
cessively shy, others lazy, and still others lacking in adapt- 
ability. On the whole, the average child of our group is 
perhaps slightly better adapted to school life than is the 
strictly average gifted child of the same degree of bright- 



34 TRAITS OF GIFTED CHILDREN 

ness. It is improbable, however, that the general character 
of our group would have been significantly different from 
what it was, had it included all the cases that were over- 
looked. 

It has not yet been possible to make an analysis of the 
nominees who failed to qualify for the gifted group. In all, 
approximately 20,000 nominations were made in the main 
search. Many of these were duplicate nominations, but it 
is probable that the number of different children nominated 
was in the neighborhood of 10,000. Practically all of these 
were given some kind of intelligence test, probably 8,000 of 
them the National. Many of the test blanks were scored 
only far enough to show that the required standard had not 
been met. It is hoped that sometime it will be possible to 
complete the scoring and to compare the relative showing 
made by various groups of nominees. The results of such 
a comparison should prove very instructive. 

SPECIAL ABILITY CASES 

The main search was not calculated to bring to light spe- 
cial ability cases, and a supplementary search was therefore 
planned for this purpose. It was thought that this procedure 
would probably net a few cases of very superior general 
intelligence who had passed with their teachers as having 
special talent only. It was hoped, however, to locate a num- 
ber of genuine special ability cases for study and follow-up. 
Special teachers of art, music, manual training, domestic 
science, and agriculture filled out the nomination blank 
shown on the next page. 

This blank was used for about eight weeks, but it netted 
so few cases showing evidence of real talent that it was 
abandoned. In general intelligence the special ability sub- 
jects averaged 114 IQ. Among the most marked cases of 
musical ability were three children (all boys) who quali- 
fied for the regular gifted group. Ten others, three boys 
and seven girls, were purely special ability cases* Some of 
the latter were located as a result of inquiries among pri- 
vate music teachers. These ten cases had an average IQ of 
122, with a range from 95 to 139. 

The search for special talent in art was especially disap- 
pointing in all three centers. In Los Angeles a special art 



SEARCH FOR SUBJECTS 35 

BLANK FOR THE SELECTION OF GIFTED CHILDREN 
SPECIAL SUBJECT TEACHER 

TO BK FILLED OUT AND RETURNED TO THE PRINCIPAL'S OFFICE BEFORE 

Name of teacher School (or Schools) City 

To THE SPECIAL SUBJECT TEACHER: The purpose of this blank is to locate chil- 
dren who show exceptionally superior talent in such SPECIAL LINES as music, draw- 
ing, painting, modeling, dramatics, mechanical ingenuity, use of tools, etc. Below, 
write the names of from one to three children who, of all the children enrolled in 
your special subject classes, show the greatest amount of talent. Name at least 
one pupil for each special subject you teach. Give one or even two additional 
names in case you have that many who show very exceptional talent. 

Take age into account. Of two pupils who do work of about equal quality, 
but who differ one or more years in age, the younger is probably the more talented. 

SUBJECT TAUGHT. Grade or grades Total No. of children 

1. Most talented.. .Age (yrs, and mos,) Grade School 

Is this child*s talent "very extraordinary,** "decidedly superior,'* or only 
"slightly superior?" 



2. Next most talented. Age (yrs. and. mos.) ...Grade School...- 

Is No. 2 almost as exceptional as No. 1, or considerably less exceptional? (Un- 
derline) 

3. Third most talented .Age (yrs. and mos.) Grade School 

Is No. 3 almost as exceptional as No. 2, or considerably less exceptional? (Un- 
derline) 

ADDITIONAL SUBJECT TAUGHT Grade or grades Total No. of children 

1. Most talented .Age (yrs. and mos.) Grade School 

Is this child's talent **very extraordinary," "decidedly" superior,** or only 
"slightly superior?** 



2. Next most talented..... Age (yrs. and. mos.) Grade School 

Is No. 2 almost as exceptional as No. 1, or considerably less exceptional? (Un- 
derline) 

3. Third most talented. Age (yrs. and mos.) Grade School.,..- 

Is No. 3 almost as exceptional as No. 2, or considerably less exceptional? (Un- 
derline) 

class is maintained for the 50 most promising pupils in the 
city. These pupils meet twice a week, in two classes of 25 
each, to be given instruction by the best art supervisors in 
the city. The selection of pupils for this privilege involves, 
first, nomination by regular teacher to art teacher, then 
recommendation from art teacher to art supervisor, and, 
finally, recommendation from supervisor to head supervisor. 
Of the 50 children attending these classes, the 6 boys and 9 



36 



TRAITS OF GIFTED CHILDREN 



girls most promising as judged by the art supervisors, were 
examined. Not only was the average IQ low (109, range 79 
to 133), but with two or three possible exceptions there was 
little evidence of anything more than "copying" ability. It 
must be admitted, however, that satisfactory measuring 
methods were not available. The children themselves, with 
few exceptions, were looking forward to other careers than 
art There is considerable evidence that exceptional talent 
in painting or drawing is less likely than musical ability to 
appear precociously. Every city has its musical prodigies 
who give recitals, but it is said that in the history of New 
York City there has been but one public exhibition by a child 
artist. It will be interesting in the follow-up study to note 
how many of the 643 children in our main experimental 
group later show well-marked specialized talent. 

SEARCH IN OUTSIDE SCHOOLS 

In the outside schools a nomination blank similar to the 
one given above was used except that the teacher was asked 
to nominate only the brightest, second brightest, and young- 
est in her class. The nominees of grades 1 and 2 were then 
given a Stanford-Binet test, those of grades 3 to 8 a National 
Intelligence Test (Scale A or Scale B), and those of grades 
9 to 12 a Terman Group Test. The standard for acceptance 
on the National was as follows : 

Score 

73 
82 
90 
97 
104 
111 
118 
125 
132 
138 
144 
149 
154 
159 
163 
167 





Age 




7 bi 


at nc 


>t yet 7% 


7% 




" 8 


8 




" 8% 


8% 




" 9 


9 




" 9% 


9% 




" 10 


10 




1 10% 


10% 




* 11 


11 




* 11% 


11% 




* 12 


12 




12% 


12% 




* 13 


13 




" 13% 


13% 




" 14 


14 


4 


" 14% 


14% ' 


i 


" 15 



SEARCH FOR SUBJECTS 



37 



The minimum T. G. T. scores for high school pupils 
were as follows: 



Below 12 

12 but not yet 



12% 




* 


* 13 


13 




< 


1 13% 


13% 




t 


' 14 


14 




t 


' 14% 


14% 




i 


4 15 


15 or 


ibov 


e 





Score 

135 
145 
155 
165 
174 
183 
190 
195 

Although the outside surveys netted many interesting 
subjects whom it will be well worth while to follow up, their 
immediate results were disappointing because of the impos- 
sibility of controlling conditions. The outside cases selected 
on the basis of a Binet test numbered 356; and on the basis 
of a Terman Group Test or National Intelligence Test, 378; 
all outside cases, 734. The results for these are not very suit- 
able for statistical treatment because of certain departures 
from instructions, and because not all of the subjects quali- 
fied on the same kind of test. Also, the supplementary data 
are far less complete than they are for our main group. 
Some of the most important data for 309 of the outside cases 
located in the high school survey are summarized in Chap- 
ter XIX. 



CHAPTER III 

COMPOSITION OF THE GIFTED GROUP 1 

By May 1, 1924, our files included 1,444 subjects, classified 
as in Table 2. 

TABLE 2 

CLASSIFICATION OF THE GIFTED GROUPS 
/. Main Experimental Group: Boys Girls Total 

1. Up to standard and parents cooperating 356 301 657 

2. Up to standard but inadequate coopera- 

tion from parents 5 7 12 

3. Somewhat below standard but included 

for special reasons 8 5 13 

4. Deceased 112 



Total, Group I 370 314 684 
//. Outside Binet Group: 

1. Parents cooperating 130 120 250 

2. Parents not cooperating 67 39 106 

Total, Group II 197 159 356 
///. Outside High School Group: 

1. Parents cooperating 201 98 299 

2. Parents not cooperating 55 23 78 

3. Deceased. 101 



Total, Group III 257 121 378 
IV. Special Ability Group: 

1, Parents cooperating 8 13 21 

2. Parents not cooperating 235 



Total, Group IV 10 16 26 

Grand Total 831 613 1,444 

In addition to the above 1,444, there are 211 outside cases 
(103 boys and 108 girls) in our files who failed by a small 
margin to qualify but are being followed up. 

^Written with the assistance of Florence L. Goodenoxigh. 

39 



40 TRAITS OF GIFTED CHILDREN 

Group I is by far the most important group, as it is com- 
posed almost entirely of subjects who were located in the 
main survey and for whom we have collected the most ex- 
tensive data. It consists of 684 subjects. However, what is 
referred to throughout this volume as the "main experimen- 
tal group" included only 643 of these subjects. This restric- 
tion of numbers is due to the following facts: (1) 33 of the 
684 cases were siblings of the original 643 and were either 
tested for the first time a year after the original survey was 
made or were later transferred to the main experimental 
group from other classifications. (2) Two boys with IQ's of 
111 and 114, twins of qualified subjects, were included in 
order to avoid possible jealousy. These cases have not been 
included in any of the statistical reports. (3) Parental coop- 
eration was refused in 6 cases. These three groups account 
for 41 cases, which, subtracted from 684, gives the 643 cases 
of the main experimental group. 

Group II includes 128 old cases that had been located be- 
tween 1911 and 1921, and 228 that were located in 1921- 
1922 by the help of volunteer assistants in cities of Califor- 
nia other than those covered in the main search. It is the 
outside pre-high school group, and was selected for the most 
part on the basis of National and Binet tests. The data which 
have been collected for this group have not been summa- 
rized, largely because they are so incomplete. 

Group III is the result of a cooperative survey carried on 
by principals of 95 high schools in California, and was se- 
lected chiefly by the use of the Terman Group Test, Chap- 
ter XIX has been devoted to a summary of some of the data 
collected for this group. 

,The nature of Group IV has been indicated in the pre- 
ceding chapter. 

COMPOSITION OF MAIN EXPERIMENTAL GROUP 

Table 3 shows the distribution of the unconnected IQ's of 
the main experimental group of 643 subjects by age and sex. 
It is seen that the largest numbers of cases are found in the 
age groups 8 to 12, inclusive, and that the mode is at 10* 
This is due for the most part to the fact that the search was 
carried out chiefly in grades 3 to 8. Many bright children 
of 7 years or less have not yet reached the third grade, while 



COMPOSITION OF THE GIFTED GROUP 



41 





3< 


3 2 


TH CM CO 


t^ OO 

IO G 


CO 

* 


OS 

OS 


^ OO 
CO 


CO 




o r 


2 ^ 






TH 


TH 




CO 




H p 


c/2 




















^-i ^ ^ -^ ^ |>, ^ 


CO 10 


CM 


CO 


CO CO 


TH 






O 


TH 


CM CO 


t-* 


OS 


CO 




















CM 




CC 


















o 




CM' oo -^ oo 


^ CO 


TH 


CO 


44 CM 


CM 






CQ 


TH TH 


CO CO 


^ 


o 

TH 


CO 


10 
CO 






O 








CM 


r- ^ 


CO 

TH 




CO 


















TH 


ffi 






TH 


CO 


lO TH 


CO 
TH 


04 




CM 


O 




TH TH 


CO 


CO 

TH 


CO TH 
TH 


00 
CO 


c: 


TH 
















O 









CM 


CO 


CO 
TH 


TH TH 
TH 


CO 
CO 


Q 


















1 




O 


00 


CO *O 


CO 

TH 


1O 

TH 


"* 


OS 


O 


TH 
















2 




CQ 


TH CM 


^f CO 

TH 


tO 

TH 


TH 


oo 


o 

CO 


o 




TH 


O 




TH 


CM 


TH 




iO 






_Q 


TH CO CO 


*O CM 


CO 


CM 


CM 


C-. 


W 




^4 




TH TH 


TH 


CM 




"* 


O? 


















^ 




O 


TH TH CM TH 


^ OS 





TH 


TH 


CM 


B 


0> 
















O 






TH CO lO 


co co 


TH 


CO 


CO 


TH 






CQ 




TH 


TH 


TH 




co 


O 

u 




C? 


rH *""* CM CM *O 


KO CO 


OS 


CM 

TH 


CM 


CM 


fc 


00 
















P 




PQ 


TH CO CO CO 


CO TH 

TH 


OS 


TH 
TH 




3 


o 


















55 




O 


TH 


TH TH 


OO 


OO 


TH 


o 

CM 


P 




CQ 


CM CO 


TH nO 


> 


* 


TH 


CO 
CM 


CO 




O 


TH CM TH 


TH * 


TH 


OS 


TH 


O 
CM 


M 

O 


CO 




















ff 




CO CO 


CM 


I> 


TH 


CO 

TH 






O 




CM 


CM 


CO 




t- 




10 




















CQ 


TH TH 


TH 


* 


CM 


TH 


TH 

TH 




1 


O 


TH 


rH 




CO 


TH 


CO 






PQ 




CM 


CM 


CM 


CM 


OO 
















CO 






1 


O? 

HH 


oososoooot^t^coco 


10 


IO 


o 


10 O 
CO CO 


1 



42 



TRAITS OF GIFTED CHILDREN 



many of the brightest 12 year olds and nearly all of the 
brightest 13 year olds are in the high school. The decrease 
in the number qualifying after age 11, and the relative in- 
frequency of very high IQ's among the older subjects who 
did qualify, is in part due to the inadequacy of the Stanford- 
Binet for the older gifted, although some allowance was 
made for this ty lowering somewhat the standard for qual- 
ification for subjects 12 years old or older. 

An attempt has been made to correct the IQ's to corre- 
spond to what they would have been had the scale been 
more nearly adequate in the upper range. Various methods 
of making this correction were tried empirically until one 
was found which seemed satisfactory. 1 The correction used 

*By the first method attempted, the gifted children were divided into 
groups according to the basal year. For each basal year-group of subjects 
the average number of months of mental age earned in each year-gn>up of 
tests above the basal year, was calculated. Curves were plotted and 
smoothed. The ratio of the mean total increment to the basal year was 
next calculated. It was found, however, that to take this ratio alone as a 
basis for correction gave very questionable results in individual cases. 

The next method tried was that of plotting individual tests to see 
whether the amount of scatter tended to preserve a fairly constant ratio 
for successive years. In general, this was found to be the case. The results 
of a test in which the child has "worked himself out'* may be represented 
graphically by the following figure in which the point A indicates the 
basal year. 



If the scale is inadequate to show the true mental level, it may be indi- 
cated thus: 



A B 

The point G indicates the limit of the scale; A, the basal year; and B some 
intermediate year. Accordingly, the ratio of the number of months gained 
beyond any point after the basal year to the mean of the months scored in 
the two immediately preceding years was taken as the basis for correction. 
This was found to give more consistent results than the first method (ratio 
of total increment to basal year). By applying this ratio and smoothing, 
the correction table given above was derived. 



COMPOSITION OF THE GIFTED GROUP 43 

involved the following additions of months to the mental 
age score for those passing various numbers of tests out of 
the total of twelve tests in year groups 16 and 18 : 

Tests passed in 16 and 18 567 8 9 10 11 12 
Number of months to add 3 6 9 12 15 18 21 24 

Applying this correction we have the distribution shown in 
Table 4. 

It is seen that the corrected quotients reduce consider- 
ably the age differences with respect to various grades of 
superiority. The proportion of IQ's as high as 170 at vari- 
ous age levels in Table 4 is as follows : 

Age 2 to 5, 2 of 32 cases, or 6.2% 
Age 6 and 7, 4 of 82 cases, or 4.8% 
Age 8 and 9, 16 of 189 cases, or 8.5% 
Age 10 and 11, 19 of 240 cases, or 7.9% 
Age 12 and 13, 2 of 100 cases, or 2.0% 

Only at age 12 and 13 is the difference very marked, and 
here the drop may be explained as due largely to the fact 
that many of the brightest children of these ages have been 
promoted to the high school. The mean corrected IQ by 
age is as follows : 

2 to 5 6 and 7 8 and 9 10 and 11 12 and 13 
148.75 149.45 151.95 153.9 149.1 

Accordingly, when we allow for the known inaccuracies of 
the intelligence scale used, the frequency of high-scoring 
children shows little tendency to increase or decrease in the 
total age range covered. For this reason it seems justifiable 
to throw all the ages together for a total IQ distribution. 

It will be noted that 21 cases were included whose cor- 
rected IQ's are in the interval 135-139, and one who is in the 
interval 130-134. The large majority of these 21 cases are 
within a point or two of IQ 140. Some of them were included 
because of indications that the Binet score, due to the condi- 
tions of the examination, was lower than it should have 
been; several of them because they were sibs of children 
who had already qualified; and a few because of special in- 
terest which attached to their cases. The number of ex- 
ceptions made is, of course, too small to affect appreciably 
the general character of the experimental group or to in- 



44 



TRAITS OF GIFTED CHILDREN 





13 


38 


TH cMCMOooor>.co**<Hti 

TH CM CM ^t< CD CO 


o 


O TH TH 
CO CM 


CO 




*- 




TH 


TH 


TH 


CO 




H 


CO 


TH CM TH CM CO OS O r-- tO 0S 


<* 


1C O> rH 


TH 




3 


O 


TH TH CO ^ 


^ 


** 


CM 




o 




THOOCMTHt>.cDcsiuri 


CO 


iO CM 


CM 




H 


CD 


TH TH CM CM 00 


I> 


OO TH 


co 








TH ^ 


co 


10 


CO 






O 








TH 




CO 














TH 


PQ 


CO 


^ 




CO 

TH 


ft 




O 


TH CM *O O 


CO 

TH 


CO TH 


oo 

CO 





CM 












e 


TH 

















CQ 


CM <0 *> 


TH 
TH 


O> TH 


CO 


o 














B 




O 


-<?< CM O> TH 
CM CM TH 


TH 

TH 


CO CM 


s 


o 


TH 

TH< 
















/v> 


- , j S| <^< QQ ^J (<>. 


CM 


CM TH 


o 


2 






TH 


TH 


TH 


CO 


i 




O 


TH TH CO TH CO JO TH 

TH 


TH 


O> TH 


s 


ft 















o 


















PQ 


"* TH CM oo r- t> r- 


o 

TH 


0> CM 


J> 


0? 














HH 




O 


TH TH CM CM l> <O 


oo 


<** TH 
TH 


CM 


1 


OS 
















PQ 


TH CM CM HT CO CO 

TH 


TH 
TH 


CO CO 

TH 


TH 
CO 


1 







TH CO TH CM "tf 4 lO CO 


O 


CM CM 

TH 


CM 




oo 

















PQ 


TH CO CO OO CO TH 

TH 


<y> 


TH 


9 


55 














g 




O 


TH TH TH 


oo 


CO TH 


o 

CM 


P 


* 












s 




PQ 


CM CO TH U3 


I*** 


r- TH 


CO 















CM 


M 

Q 




O 


TH CM TH TH ^ 


TH 


CO TH 


o 

CM 




CO 
















PQ 


CO CO 


CM 


r~ TH 


CO 

TH 






O 


CM 


CM 


CO 


t** 




O 
















PQ 


TH TH TH 


*> 


CM TH 


TH 

TH 




I 1 





^ 




CO TH 


<o 




CM 


CQ 


CM 


CM 


CM CM 


CO 












CO 






o 

to 


O 

H-f 


O O> O> OO OO t>* l> CO CO kO O 


*H 


O W3 

^r CO CO 

<_4 wf i~t 


1 



COMPOSITION OF THE GIFTED GROUP 

TABLE 5 
IQ DISTRIBUTION FOR AGES COMBINED 



45 



Unconnected 



Corrected 



IQ 


B 


G 


Total 


B 


G 


Total 


200 














1 


1 


195 




















190 





1 


1 





2 


2 


185 





1 


1 


1 


1 


2 


180 


2 


1 


3 


8 


2 


10 


175 





4 


4 


2 


6 


8 


170 


8 


4 


12 


11 


9 


20 


165 


14 


7 


21 


17 


10 


27 


160 


18 


14 


32 


26 


17 


43 


155 


34 


23 


57 


29 


35 


64 


150 


63 


35 


98 


85 


49 


134 


145 


71 


72 


143 


76 


74 


150 


140 


106 


93 


199 


85 


75 


160 


135 


34 


30 


64 


12 


9 


21 


130-134 


2 


6 


8 





1 


1 


Total 


352 


291 


643 


352 


291 


643 


Mean 


148.635 


147.825 


148.268 


151.418 


151.210 


151.33 


S. D. 


8.8605 


9.5525 


9.189 


9.858 


10.5665 


10.186 



validate conclusions which are drawn with respect to the 
mental and physical traits of children of IQ 140 or higher. 

COMPOSITION OF GROUP II 

Table 6 gives the IQ distribution of Group II, subjects 
who were selected by volunteer assistants on the basis of 
Binet tests given in California cities not covered in the main 
survey. Of these, 128 are cases which have been under ob- 
servation for several years and about 75 are from other 
states than California. Several have been tested more than 
once. When more than one test score was available the IQ 
that was entered in Table 6 is that from the test which was 
most complete. In a few instances where the tests seemed 
to be of about equal validity, the mean of the IQ's was 
taken. 

Figure 1 shows the distribution of 999 gifted cases (643 
in the main experimental group plus 356 in Group II) in 
comparison with the IQ's of the 905 unselected children on 
whom the Stanford-Binet test was standardized. The "un- 
corrected" IQ's are used in Figure 1. 



46 



TRAITS OF GIFTED CHILDREN 



W 1 "CJ ' 
g B 

c/> O ' j^ 


CM tO "tf iCSt OO CO t** CD iO C^ CM ^ 

T-t T- CO UO OO CM Y- 

CNCOCMCOO^^OOt-|gtO 


CO 
UQ 


H CQ 


CM CO Y"* CO irt l>* CM CO l> xO O "^f 

t-4 T ( 


s 


*"* CQ 
co 
^ CQ 


^-4 CO CO 
CM CM "tf 


oo 


O 

CM 

1? ^ 


Tf jQ OO 

csi CM r- oo 


t^ 


O 


T^ ^-t r-, -r-i CO ^f XO 


CO 


8 " 


T-< CM ^c* CM O> --i 


CM 


B o 

O 

ftl ** 
P m 




T-CO^fCM'*-< > ^*OO<MCM 


CM 

CO 


3 o 
eg S o* 
5 1 


T-< CO CM ^ 1O C 


CM 

O 
CO 


1 ^ 

1 

o 

o 


THT-I CNTf^^OOO* 
CM "^< CM <O '"<* CO l> 


<M 
CM 


00 t*i 
^ W 


^ TiT-ir-< ^CMCO 


00 


O 

CO 

CQ 


^ YH CM 01 *O T-I 


CM 


O 


rH TK CO CM <r-4 


oo 


CQ 


,- T-< r-i <M "^ *- 


o 

rH 





T-t -r-t CM CM CO CM 





* PQ 


CM T* r- CM CO 


o 


-t c* 


CO 

ooo,ooo3 

Oi O> OO CO * C^ %O CO IO iO *tt* "^t< CO CO 


3 

O 



COMPOSITION OF THE GIFTED GROUP 



47 



FIGURE 1 
IQ DISTRIBUTIONS FOR 999 GIFTED AND 905 UNSELECTED CHILDREN 



40% 
3O% 
20% 
10% 



I.Q. 55-65 65-75 75-65 85-95 95 -W5 WS-T15 tt5-lE3 125-135 135-W5 145-155 155M65 165-175 175-185 185-195 




COMPOSITION OF GROUP III 

Table 7 gives by age and sex the Terman Group Test 
scores of 370 of the 378 subjects in Group III, the outside 
high school group, who qualified on this test. 

COMPOSITION OF GROUP IV 

This group is composed of the special ability cases that 
failed to qualify for any of the other gifted groups. It does 
not include cases of undoubted special ability that were 
able to qualify on the basis of the IQ. The group is as 
follows : 



With artistic ability 

Sex Age IQ 

G 11 133 

G 13 127 

B 13 126 

B 13 126 

G 12 122 

G 13 115 

B 14 114 

B 13 110 

B 17 100 (little English spoken) 

B 12 92 

G 10 91 

G 15 (T. G. T. score 89) 

G 14 101 

G 14 87 

G 15 79 



With musical ability 

Sex Age IQ 

B 5 139 

G 10 136 

G 10 132 

B (not re- 131 

corded) 

G 12 125 

B 9 121 

G 12 120 

G 13 115 

G 10 110 

G 7 95 

With mechanical ability 

B 8 108 



48 



TRAITS OF GIFTED CHILDREN 







c5 


COOO'^COTHCftCOtOCMCM'^CM'^TH TH 


cn 






'S 


TH CM CM TH 


TH 

TH 






-*-* 

o 










fcH pQ 


THCOOCOTHTHTHTH 


TH 

vC 

CM 






o 


TH 


TH 






OS > 
TH O CQ 


CM CO vo 


O 

TH 




CM 


O 


CO r-( 


^ 




o 


CO 








05 

O 


TH 

CQ 


TH CM *O |> 


TH 




o 
o 


O 


TH CM CO CM 


TH 

TH 




W 


*> 











*"* CQ 


CM ^ t- ^ TH 

CM TH 


oo 




w 










2 




C? 

CO 


*o o o> 

TH 


CM 




w 
S 


CQ 


CM i> o jur^ 


co 


t^ 


O 


O 

IO 


TH co CM >^ TH 


TH 

TH 


3 

5 




C/3 


TH 

CQ 


CO C* 1 * CD CM 


CO 


H 


S 




^ ^ TH 


CM 







O 




CM 




CJ 


. H 








co 


TH 








|? 


CQ 


TH co ^ o i> r* CM 

TH 


CO 




H 


O 


THCMCMCOCO^^TH 


CM 




o 

cc 


TH 

CQ 


CMcOTfiOO>COCOCM^f 


CO 















55 


O 


CM CM CM THCMCOT-tCM 


fcO 




< 


cq 




TH 




I 


^ CQ 


WWNCO^W^^r. 


CM 

CM 






O 


TH TH CM TH TH 


CO 






^ W 


TH TH CM TH 


id 






H 


oo 








. 


TH 


-H 






O t* 










to o 


t5 **!? IO CD W5 CD *O CD )uC!^ ^^ *ft <*|*> iy^> CD xO *^* i^O ^j 


*o 






< H co 


CMCMCNOlTHTHTHTHTHTHTHTHTHTHTHrHTHlTH 


c** 



COMPOSITION OF THE GIFTED GROUP 49 

SEX RATIO 

In all of the groups except the special ability group, the 
boys are considerably more numerous than the girls, not- 
withstanding the precautions that were taken to avoid sex 
preference in the methods of search. In the main experi- 
mental group there are 352 boys to 291 girls (54.7 per cent 
boys and 45.3 per cent girls), a ratio of 121.0 to 100. The 
S.D. of the proportion is 1.96. 1 Of 33 cases later added to the 
main experimental group, there were 11 boys and 22 girls. 
If we include these, the ratio becomes 363 boys to 313 girls, 
or 116.0 to 100. Turning to Group II, the outside Binet group, 
we find 197 boys and 159 girls, giving a ratio of 123.9 to 100. 
Group III, the high school group, yields 257 boys and 121 
girls, a ratio of 212.3 to 100. 

It is well known that there is an excess of male births in 
the general population. Perhaps the most extensive statisti- 
cal study has been made by Nichols, 2 who finds a ratio of 
105.5 males to 100 females in living births of Europe and 
105.9 to 100 in living births of whites in the United States. 
The excess of males is greater in still-births and less in mul- 
tiple births. The above ratios correspond closely to those 
reported for other mammals. In the case of pure Albino 
rats, for example, Miss King found for 80 litters and 452 off- 
spring a ratio of 107.3 males to 100 females. 3 The reason for 
this excess of males is not definitely known, but a possible 
explanation is that sperms which contain the male-produc- 
ing sex chromosome, being lighter by a minute fraction than 
other sperms, may travel somewhat faster and attain the 
upper reaches of the oviduct in larger numbers.* 

However, the best standard with which to compare the 
sex ratio in the gifted group is the sex ratio in the pre-high 
school population of the cities covered by the survey. This 
was found to be 104.5 boys to 100 girls. Our problem is, 

*The S.D. of a proportion ^ / .** ^ -Q , where P = the per cent of cate- 

\ N 
gory 1, Q the per cent of category 2, and N = the number of cases. 

3 J. B. Nichols: Numerical Proportions of the Sexes at Birth. Memoirs 
of Am. Anthrop. Ass'n. Vol. 1, 1907, pp. 249-300. 

"Helen D..King: The Sex Ratio in Hybrid Rats. Pub. by the Wistar 
Institute, 

*See T. H, Morgan: Mechanism of Mendelian Heredity. 1923, pp. 
134-140. 



50 TRAITS OF GIFTED CHILDREN 

therefore, to explain the difference between this ratio and 

those found for the gifted groups, the latter ranging from 

1 1 fi 5^1 ^ 

TAA to TATT Four hypotheses will be examined. 

lUU 1UU 

1. Biased selection. It is possible that the method of se- 
lection may have favored the boys, although in view of the 
fact that nominations on the basis of estimated intelligence 
were in the vast majority of cases made by women teach- 
ers, one would hardly expect this to be the case* The 
nomination blanks filled out by teachers were examined for 
evidence of sex preference. Owing to the difficulty of iden- 
tifying sex of a nominee by the name, some inaccuracy is 
introduced, and examination was therefore not carried out 
for the 6,000 or more blanks secured, but only for those filled 
out by the teachers of children who qualified for the gifted 
group. It is especially important to know whether these 
particular teachers showed a large sex preference in their 
nominations. 

Of the original 643 cases (352 boys and 291 girls), 257 
boys and 190 girls had been located as a result of nomina- 
tions. The others were discovered in various ways, as ex- 
plained elsewhere. The figures 257 and 190 give a ratio of 
135.3 boys to 100 girls. The blanks on which these children 
were nominated contained nominations of 1,010 boys (765 
in addition to those qualifying) and 920 girls (738 in addi- 
tion to those qualifying). The figures 1,010 and 920 are in 
the ratio of 109.7 boys to 100 girls. Omitting those qualify- 
ing, the ratio is 103.7 boys to 100 girls, which is almost ex- 
actly the same as the ratio for the entire school population 
in the cities covered. One could hardly conclude that the 
excess of boys in our group is due to the bias of teachers in 
making nominations. Even a large excess of boys nomi- 
nated would not be proof of bias. The fact that the sex ratio 
of nominations on these blanks was 109.7 boys to 100 girls, 
whereas the sex ratio of those qualifying from these same 
nominations was 135.3 boys to 100 girls, would seem to free 
the teachers of the suspicion of bias- 
There remains the question whether the Stanford-Binet 
test, which was relied upon for the final selection, is more 
favorable to boys than to girls. It is not possible here to 
review the numerous investigations that have been reported 



COMPOSITION OF THE GIFTED GROUP 51 

on this point in the literature of mental tests, nor is it nec- 
essary. The results have shown fairly consistently that, age 
for age and grade for grade, girls do fully as well on this test 
as boys. 

Private schools, of which there are a considerable num- 
ber in the cities canvassed, were not included in the survey. 
It is, of course, possible that such schools enrolled more 
gifted girls than gifted boys. We consider this unlikely, but 
the facts are not available. 

2. Sex ratio in families of gifted. At the time the mate- 
rial on sex ratio was tabulated, data were available for 502 
of the total 578 families which produced the main gifted 
group of 643 children. These 502 families yielded 317 gifted 
boys and 274 gifted girls, a ratio of 115.7 to 100. The total 
number of children in the same families was 655 boys and 
548 girls, giving a ratio of 119.5 to 100. The ratio among the 
sibs of the gifted was 123.35 to 100, 1 which is appreciably 
higher than for the gifted themselves. 2 It appears, there- 
fore, that the factor which operates to give an excess of 
boys among the gifted affects no less strongly the sibs of 
the gifted. It has been suggested that superior vigor or 
vitality of parents favors maleness of offspring, and that 
this factor might at the same time exert a favorable influ- 
ence upon the nervous structure and mental development 
of the offspring. There is a certain amount of indirect evi- 
dence which supports this hypothesis. Riddle 3 found that 
in the case of pigeons, in which the determiner for sex is 
carried by the female (it is carried by the male in mam- 
mals), if the female is stimulated by removal of eggs from 
the nest to keep on laying, the eggs later produced result in 
an excess of female offspring. By analogy, one might infer 
that in the case of human beings, superior vigor of fathers 

*Not calculated by simple omission of the gifted themselves. To do 
this would introduce a statistical error, since it would mean giving least 
weight to those families producing most gifted children. The method fol- 
lowed was that of counting the siblings of each gifted child regardless of 
the number of gifted produced by the family* This involves duplications 
in the case of those families producing more than one gifted child and 
means weighting the sex ratio in each family in proportion to the num- 
ber of gifted children produced by it. 

*The gifted having an IQ of 160 or higher had 86 brothers and sisters, 
a ratio of 112.6 to 100. 

Oscar Riddle: Sex Control in Pigeons. American Naturalist, July, 1916. 
Vol. 50, pp. 385-410. 



52 TRAITS OF GIFTED CHILDREN 

would result in an excess of male births. It need hardly be 
said that analogical reasoning in the biological field has no 
value except in so far as it suggests investigation. It is true, 
however, that the medical and anthropological data which 
we have secured indicate that gifted children come from 
families of more than average vigor. As will be shown in 
later chapters, the children themselves are well-nourished 
and average above normal in height and weight; infant 
mortality among the sibs has been very low; and the grand- 
fathers rate considerably above average in longevity. 

In a study of longevity, 150 families were selected which 
seemed to show the greatest tendency to long life. The se- 
lection could not readily be made on a purely objective 
basis, owing to the presence of so many variables, such as 
age of relatives, number in the family, number of deaths 
that had occurred, occupation, etc., but there is no doubt 
that the method used gave on the whole a group of families 
of superior longevity. Possible bias was ruled out by having 
the selections made by an assistant who had no knowledge 
of the use to be made of the data. In the 150 families so 
selected there were in all 253 children; 136 boys and 117 
girls. This gives a ratio of 116.2 to 100, as compared with a 
ratio of 119.5 to 100 for the children of the entire group of 
families for which we have data. For the 253 children of 
long-lived families there were 552 uncles and 509 aunts re- 
ported, a ratio of 108.2 to 100, which is very close to that 
found in the general population. 

It seems, therefore, that whatever the factor involved, 
it operates to give an excess of males only among the gifted 
children and their sibs, not among their relatives. In line 
with this is the fact that 478 of CattelFs American Men of 
Science had 716 sons and 668 daughters, a ratio of 107.2 to 
100, or almost exactly the same as the ratio for the general 
population. The facts presented above offer no evidence 
that the general stock which produces gifted individuals is 
characterized by excess of male offspring. 

3. Differential death rate of embryos. Our figures for 
sex ratio among the gifted and their sibs are based upon 
living births. Nichols and others have reported some evi- 
dence that the proportion of males may be higher in mis- 
carriages. Hospital reports have in some cases shown a 
third as many miscarriages as living births. The number 



COMPOSITION OF THE GIFTED GROUP 53 

may be considerably greater than this, as it is impossible to 
secure accurate data on the mortality of embryos within 
the first few weeks after conception. If the ratio of males 
to females for all conceptions were 120 to 100, and if the 
mortality rate of male and female embryos were 30 per cent 
and 20 per cent, respectively, then the sex ratio of living 
births would be 105 to 100 (or approximately that which 
obtains for the general population). Assuming the same sex 
ratio for all conceptions, but a mortality rate of 15 per cent 
for male and 10 per cent for female embryos, then the sex 
ratio of living births would be 113.3 males to 100 females. 
It is evident, therefore, that a reduction of 50 per cent in 
miscarriages, using this term in the broad sense, might af- 
fect considerably the sex ratio of living births. 

We find that in the families reporting one or more mis- 
carriages there were, of living births, 183 boys and 160 girls; 
a ratio of 114.3 to 100, or 53.4 per cent of males and 46.6 per 
cent of females. In the families reporting no miscarriages 
there were 466 boys and 380 girls; a ratio of 122.6 to 100, 
or 55.1 per cent of males and 44.1 per cent of females. The 
S.D. of the first proportion is 2.69; of the latter, 1.71. The 
excessive proportion of boys in the families reporting no 
miscarriages may very well be significant. If mothers of 
the gifted group on the whole have excelled mothers of the 
generality in the ratio of live births to conceptions, the ex- 
cess of gifted boys would readily be accounted for. 

4, Sex difference in variability. The most common ex- 
planation of findings such as we are here concerned with 
is that the human male is more variable than the female. 
However, the mental test data bearing on sex variability are 
so inconsistent that it would be hard to say which way the 
weight of evidence inclines. On the hypothesis of sex dif- 
ference in variability one would expect to find the highest 
intelligence scores in our gifted group earned by boys. This, 
however, is not the case in all groups. In the main experi- 
mental group the three highest IQ's (corrected) were earned 
by girls. Both in Table 3 (uncorrected IQ's) and Table 4 
(corrected IQ's) the means for the sexes are almost identi- 
cal. The variability of the girls is in each case slightly 
greater than that for boys, but the difference is too small 
to be considered significant, the difference between the S.D's 
in the case of the corrected IQ's being only 0.82 times the 



54 TRAITS OF GIFTED CHILDREN 

S.D. of the difference, which is 1.06. The proportions of 
each sex found at or above various levels are for the cor- 
rected IQ's as follows: 

160 or 170 or 180 or 190 or 
above above above above 

Boys 18.5% 6.2% 2.6% 0,0% 

Girls 16.5% 7.2% 2.0% 1.0% 

However, except for the non-excess of boys in the highest 
IQ ranges, the facts we have presented are in harmony with 
the hypothesis that exceptionally superior intelligence oc- 
curs with greater frequency among boys than among girls. 
That the sibs of the gifted show even a greater excess of 
boys than do the gifted themselves may seem at first thought 
to argue against this hypothesis. Actually it supports it. If 
children are selected for a given trait, and that trait is more 
common in boys than in girls, it follows that a greater pro- 
portion of families consisting of girls only will be missed, 
with the result that the sibs of the children selected for the 
trait will also show an excess of boys. It can be shown, for 
example, that if the true sex ratio in a general population 
is 1 to 1, and the trait in question occurs in one-half of boys 
and one-third of girls, the families of children selected for 
the trait will yield a sex ratio of 111 boys to 100 girls, not 
100 to 100. 1 

The true cause of the sex ratio found can not be deter- 
mined from our data. It may be either variability or the 
differential death rate of embryos. Both of these factors 
may be involved and possibly others. Biased selection due 
to the method of nomination and testing is probably not 
responsible. 

lf The author is indebted to Professor C, H. 0anforth f of the Depart- 
ment of Anatomy, Stanford University, for pointing out this selective 
factor and for the deviation of the apparent ratio in this illustration. 



CHAPTER IV 

RACIAL AND SOCIAL ORIGIN l 

RACIAL ORIGIN 

The chief data on racial origin consist of replies by the 
child's parents to the following question asked in the Home 
Blank regarding each of the child's grandparents : Father's 
father's descent (for example, English and Scotch, German, 
Dutch and French, Russian Jewish, etc.) 

The question was answered for 85 per cent of the grand- 
parents of those for whom the Home Blanks had been re- 
turned at the time tabulations were made. In tabulating 
the replies, a credit of two points was allowed for each 
grandparent, or one point for each great-grandparent, mak- 
ing a total of 8 points for each child. The points were then 
totaled by nationality or race for each grandparent and re- 
duced to per cents, as shown in Table 8. 

TABLE 8 

RACIAL ORIGIN OF MAIN GIFTED GROUP 
(Per cent of grandparents of each racial or nationality group) 

Per cent of Per cent of 

Racial Stock total Racial Stock total 

English 30.7% Austrian 1.3% 

German 15.7% Norwegian 0.9% 

Scotch 11.3% Danish 0.9% 

Irish 9.0% Japanese 0.6% 

French 5.7% Swiss 0.6% 

Russian Jewish 3.8% Spanish , 0.3% 

German Jewish 1.8% Bohemian 0.3% 

Polish Jewish 0.8% Russian 0.3% 

Roumanian Jewish 0.2% Hungarian 0.3% 

French Jewish 0.2% Roumanian 0.3% 

Lithuanian Jewish 0.1% Flemish 0.3% 

Austrian Jewish 0.1% Armenian 0.3% 

Bohemian Jewish 0.1% Portuguese 0.3% 

Jewish, not classified 3.4% Alsatian 0.1% 

Total Jewish 10.5% Negro 0.1% 

Scotch Irish 2.8% Indian 0.1% 

Swedish 2.5% Mexican 0.1% 

Italian 1.4% Syrian 0.1% 

Welsh 1.4% Icelandic 0.1% 

^Written with the assistance of Florence L. Goodenough. 

55 



56 TRAITS OF GIFTED CHILDREN 

It is unfortunate that no very reliable data on the rela- 
tive frequency of the different racial stocks represented in 
the cities covered by the main survey are available. How- 
ever, even without such comparative figures certain items in 
Table 8 stand out as significant. The percentage of Scotch 
is very high, as is also that of the Jewish groups, especially 
the Russian Jews. The English stock heads the list, but this 
is also probably true in the general population. The per- 
centage of Latin blood is very low. 

The proportion of Jewish blood, as it is reported, is 10.5 
per cent of the total. The actual amount is probably greater 
than this, as there is reason to believe that the presence of 
Jewish blood has in some cases been concealed. According 
to the estimates of a number of prominent Jewish social 
workers, the proportion of Jews in the total population of 
the three main cities covered (Los Angeles, San Francisco, 
and Oakland) is approximately 5 per cent. According to 
this estimate, the amount of Jewish blood in our group is 
about twice the expected. 

Negroes represent 2 per cent of the total of the com- 
bined population of Los Angeles, San Francisco, Oakland, 
Alameda, and Berkeley, and furnish three-tenths of one per 
cent of our gifted group (two cases). As these cases are 
both part white (exact proportion of white blood is not 
known) they account for less than three-tenths of one per 
cent of the ancestral units in Table 8. 

In regard to the absence of Chinese, it should be noted 
that the Oriental schools which the Chinese children at- 
tended were not canvassed. Tests made by K. T. Yeung of 
105 unselected Chinese children (chiefly of the ages 9 and 
11) in the Oriental school of San Francisco gave a median IQ 
of 97, with one case testing above 135. These results compare 
well with those for unselected white children in California, 
especially if any allowance is made for the language handi- 
cap. 1 

The mentality of Japanese children 2 in California has re- 
cently been very thoroughly investigated by Professor M. L. 
Darsie, who finds little evidence that the Japanese are in~ 

*K. T. Yeung: The Intelligence of Chinese Children in San Francisco 
and Vicinity. Journal of Applied Psychology, 1921, pp. 267-274. 

a M. L. Darsic: The Mental Capacity of American Born Japanese Chil- 
dren. Comparative Psychology Monographs, 1925. 



RACIAL AND SOCIAL ORIGIN 57 

ferior to California white children. Professor Darsie gave 
Binet, Beta, and Stanford Achievement tests, using unse- 
lected age groups of 10 to 15 years, to nearly half of the 
Japanese children of these ages in the entire state. The re- 
sults for the urban group gave a median IQ of about 90, with 
Q of 10.2. For 129 cases in the larger cities the median was 
99. That the Binet median was largely affected by the 
language factor is indicated by the fact that the Beta median 
was slightly above that of unselected white children. In 
agreement with this is the fact that the median educational 
quotient of the Japanese children on the Stanford Achieve- 
ment test was about 95; also the fact that the ratings given 
these children by the teachers for general quality of school 
work was fairly high. The remarkable outcome of one 
Japanese-white marriage is set forth in some detail in 
Chapter 5. 

The total population of Latin extraction in the cities 
covered is not known, but it is certainly very large in com- 
parison with the number of Latin children in our group. 
Intelligence tests of many Latin groups in America have 
yielded consistently low scores, with a median IQ usually 
between 75 and 85. Perhaps a median IQ of 80 for the Italian, 
Portuguese, and Mexican school children in the cities of 
California would be a liberal estimate. 1 How much of this 
inferiority is due to the language handicap and to other 
environmental factors it is impossible to say, but the rela- 
tively good showing made by certain other immigrant groups 
similarly handicapped would suggest that the true causes 
lie deeper than environment. 

Examination of the statistical probabilities will show 
that even a moderate difference in the mean IQ of two race 
or nationality groups is sufficient to cause very large dif- 
ferences in the proportion of individuals testing at either 
extreme. The mean IQ of California white children residing 
in urban communities is approximately 100, and the P.E. of 
the distribution is approximately 10. On the basis of a nor- 
mal distribution, the expected frequency of children testing 
as high as 140 would be 35 in 10,000, or one individual in 286. 
If the mean IQ were 90 and the P.E. of the distribution were 

1 See Kimball Young: Mental Differences in Certain Immigrant Groups. 
University of Oregon Publications, 1922, p. 103* 



58 TRAITS OF GIFTED CHILDREN 

still 10, the proportion of cases reaching 140 would be only 
3.8 in 10,000, or one in 2,632. If the mean of 90 were accom- 
panied by a P.E. of about 9, as would most probably be the 
case, the proportion reaching 140 would be much less than 
1 in 2,632. One can at least say that a mean IQ of 90 is not 
likely to produce more than a ^tenth as many individuals 
above 140 as a mean IQ of 100 will produce. 

Table 9 gives for the gifted group and for the general 
population of the cities canvassed the proportion of "native 
white of native parentage," "native white of foreign ^or 
mixed parentage," and "foreign white parentage." The fig- 
ures for the population are taken from the 1920 census re- 
turns for "males and females 21 years or over." As would 
be expected, there are no significant differences in the par- 
entage of gifted boys and gifted girls, but the gifted rank 
higher than the general population in the proportion classed 
as "native white of native parentage/' the ratio being about 
5 to 4. 

TABLE 9 

PARENTAGE OF GIFTED COMPARED WITH ADULTS OF THE 
GENERAL POPULATION 

Native white Native white 

of native of foreign or Foreign white N<*gro 

parentage mixed parentage parentage parentage 

Male Female Male Female Male Female Male Female 

Adults in Gen- 
eral pop. 39% 42% 25% 29% 33% 26% 2% 2% 

Parents of 

Gifted boys 51% 52% 26% 27% 23% 21% 0.3% 0.3% 

Parents of 

Gifted girls 48% 50% 24% 29% 27% 20% 0.4% 0,4% 

BIRTHPLACE OF FOREIGN BORN PARENTS OF GIFTED 

The birthplaces of 248 foreign born white parents of our 
gifted are given in Table 10- The figures for the Japanese 
are added for comparative purposes* The fourth column 
gives the per cent contributed by each country to the total 
foreign born white parents of our gifted; Column 5, the per 
cent contributed by each country to the total foreign born 
population of Los Angeles and San Francisco; Column 6 
shows to what extent each number in Column 4 exceeded or 
fell short of its quota as shown in Column 5. 



RACIAL AND SOCIAL ORIGIN 



59 



TABLE 10 

BIRTHPLACE OF 248 WHITE FOREIGN BORN PARENTS OF 
GIFTED CHILDREN 



(i) 



(2) 



Country Fathers 


Mothers 


Armenia 


2 





Australia 


3 


3, 


Austria 


4 


4 


Belgium 


2 


1 


Canada 


14 


15 


Denmark 


3 


1 


England 


12 


13 


France 


4 


6 


Germany 


15 


12 


Hungary 


2 


2 


Ireland 


7 


3 


Italy 


3 


2 


Mexico 


- 


1 


Netherlands 


3 





Norway 


5 


2 


Poland 


3 


5 


Roumania 


7 


7 


Russia 


35 


25 


Scotland 


3 


4 


South America 


1 


- 


Spain 


1 


- 


Sweden 


5 


5 


Wales 


1 





Pacific Islands 


- 


2 



Japan 



(3) 



Total 

2 
6 
8 

3 
29 

4 

25 
10 
27 

4 
10 

5 

1 

3 

7 

8 

14 
60 

7 

1 

1 
10 

1 

2 

248 
10 



(4) 


(5) 


(6) 


Per cent 






of white 


Per cent 


Per cent 


foreign 


of total 


of quota 


born par- 


foreign in 


among 


ents of 


general pop- 


parents 


gifted 


ulation 


of gifted 


0.8% 


0.3% 


267% 


2.4% 


0.8% 


300% 


3.2% 


2.3% 


139% 


1.2% 


0.4% 


300% 


11.7% 


8.2% 


143% 


1.6% 


2.1% 


76% 


10.1% 


8.6% 


117% 


4.0% 


3.8% 


105% 


10.9% 


11.5% 


95% 


1.6% 


1.2% 


133% 


4.0% 


9.2% 


43% 


2.0% 


12.6% 


16% 


0.4% 


10.1% 


4% 


1.2% 


0.6% 


200% 


2.8% 


1.9% 


147% 


3.2% 


1.7% 


188% 


5.6% 


0.7% 


800% 


24.2% 


6.1% 


397% 


2.8% 


2.5% 


112% 


0.4% 


0.4% 


100% 


0.4% 


1.3% 


31% 


4.0% 


4.5% 


89% 


0.4% 


0.4% 


100% 


0.8% 


0.3% 


267% 


99.7% 






0.8% 


1.6% 


50% 



In comparing the per cents in Column 6, it is necessary to 
bear in mind the small numbers in Column 3 upon which 
they are based. This applies especially in those cases in 
which the per cent in Column 5 is also small. The great ex- 
cess over the quota for Russia (Russian Jews?) and the 
enormous deficiency below the quota for Spain and Italy, 
are, however, very significant. 



60 TRAITS OF GIFTED CHILDREN 

BIRTHPLACE OF AMERICAN BORN PARENTS AND GRANDPARENTS 

OF GIFTED 

A separate space was provided in the Home Blank for 
recording the birthplace of each parent and each grand- 
parent. Replies definite enough for tabulation were received 
for 581 fathers and 583 mothers, or more than 90 per cent of 
the full number for the main experimental group and more 
than 95 per cent of the number returning Home Blanks. 
Following are the proportions of the American born and 
foreign born parents from cities, towns, and country in the 
cases for which the information was available. The classifi- 
cation of cities by population is based on the 1890 census for 
parents, and that of 1860 for grandparents. 

From From From 

Cities of Cities and Rural dls- 

10,000 or towns of tricts less 

N over 1,000 to 10,000 than 1,000 

Fathers 581 48.4% 23.7% 27.9% 

Mothers 584 52.4% 22.3% 25.3% 

Both 1165 50.4% 23.0% 26.6% 

Grandparents 639 36,3% 34.9% 28.8% 

Birthplace of Grandparents. The data on birthplace of 
grandparents are less complete and less accurate than for 
the parents. However, there were 639 American born grand- 
parents whose birthplaces as reported were classifiable. For- 
eign born grandparents were omitted from this comparison 
because of the difficulty of obtaining population data.* 
Whether we consider the parents or the grandparents, there 
is a much larger number of cases in the urban groups than 
was true for the general population at the two periods (1890 
and 1860). However, the significance of this fact is not very 
clear, since the gifted subjects whose parents and grand- 
parents we are considering were resident in cities* Figures 
based upon gifted children found in rural schools would 

*In classifying American born grandparents a case was included un- 
der "rural" if only state and county of birthplace had been reported (e.g., 
Erie Co., Pennsylvania). If state only was reported the case was thrown 
out unless additional evidence from other parts of the blank made the 
classification fairly certain. If in such a case the occupation of the grand- 
father was given as "farmer" the birthplace was recorded as rural. This 
procedure seemed justifiable in order to counterbalance the natural tend- 
ency to remember and report urban birthplaces more frequently than rural 
because of the greater importance of the places* 



RACIAL AND SOCIAL ORIGIN 



61 



undoubtedly be very different. Perhaps the point of greatest 
interest is the fact that, notwithstanding the small urban 
population in the United States in 1860, the per cent of 
grandparents of rural birth is as low as that for parents. 

The birthplaces of American born parents of our gifted 
were tabulated by state, and although the figures have little 
significance in the absence of accurate comparative data for 
the general population of the cities canvassed, they are given 
for whatever they may be worth. 

TABLE 11 
BIRTHPLACE OF AMERICAN BORN PARENTS BY STATE 







Per cent 






of Amer. 


State 


N 


born 


Alabama 


6 


0.7% 


Arizona 


3 


0.3% 


Arkansas 


2 


0.2% 


California 


253 


27.7% 


Colorado 


21 


2.3% 


Connecticut 


5 


0.5% 


Delaware 


2 


0.2% 


Dist. of Columbia 


2 


0.2% 


Florida 





0.0% 


G'eorgia 


6 


0.7% 


Idaho 


2 


0.2% 


Illinois 


69 


7.6% 


Indiana 


27 


3.0% 


Iowa 


50 


5.5% 


Kansas 


39 


4.2% 


Kentucky 


5 


0.5% 


Louisiana 


3 


0.3% 


Maine 


5 


0.5% 


Maryland 


6 


0.7% 


Massachusetts 


15 


1.6% 


Michigan 


35 


3.8% 


Minnesota 


28 


3.1% 


Mississippi 


6 


0.7% 


Missouri 


35 


3.8% 


Montana 


4 


0.4% 


Nebraska 


20 


2.2% 







Per cent 






of Amer. 


State 


N 


born 


Nevada 


5 


0.5% 


New Hampshire 


4 


0.4% 


New Jersey 


6 


0.7% 


New Mexico 


3 


0.3% 


New York 


63 


6.9% 


North Carolina 


3 


0.3% 


North Dakota 





0.0% 


Ohio 


51 


5.6% 


Oklahoma 


2 


0.2% 


Oregon 


10 


1.1% 


Pennsylvania 


37 


4.1% 


Rhode Island 





0.0% 


South Carolina 


2 


0.2% 


South Dakota 


2 


0.2% 


Tennessee 


11 


1.2% 


Texas 


18 


2.0% 


Utah 


7 


0.8% 


Vermont 


2 


0.2% 


Virginia 


6 


0.7% 


Washington 


4 


0.4% 


West Virginia 


2 


0.2% 


Wisconsin 


24 


2.6% 


Wyoming 





0.0% 


Hawaii 


1 


0.1% 



Total 



912 



99.4 



OCCUPATION OF FATHERS 

In the Home Blank, spaces were provided for reporting 
"father's main occupation at successive ages," including age 
when each occupation that had been followed was begun, 
and the number of years it was followed. The replies were 



62 TRAITS OF GIFTED CHILDREN 

sufficiently definite for 560 fathers in the main group to per- 
mit of classification of occupation. The method of classifi- 
cation employed is that used in the U. S. census report : pro- 
fessional, commercial, industrial, and public service, with 
sub-groups under each. The results are as follows for the 
"main occupation" of fathers at the time the report was 
made: 

Professional Group: Number 

Lawyers 33 

Engineers (with college degrees) 28 

Teachers (total 30) 30 

In colleges and universities 10 

High school 13 

Elementary 3 

Unclassified 4 

Physicians and surgeons 23 

Clergymen 15 

Writers 9 

Dentists 9 

Musicians 8 

Architects 3 

Inventors 2 

Other professions 5 

Total professional 163 

Proportion of all fathers 29.1% 

Commercial Group: 

Executives and managers 92 

Salesmen and insurance agents 43 

Retail dealers (small stores) 33 

Clerical workers 33 
Wholesale dealers, brokers, and owners 

of large retail establishments 19 

Manufacturers 1 1 

Druggists S 

Editors and publishers 5 

Expert accountants 5 

Total commercial 259 

Proportion of all fathers 46.2% 

Industrial Group: 

Carpenters 16 

Mechanics and machinists 14 

Tailors 7 

Painters 6 

Contractors 5 

Barbers 5 

Florists 4 



RACIAL AND SOCIAL ORIGIN 63 

Industrial Group (Continued): 

Telegraph operators 3 

Butchers 3 

Photographers 2 

Farmers 2 

Lithographer 1 

Foreman 1 

Pattern maker 1 

Landscape gardener 1 

Sea captain 1 

Baker 1 

Potter 1 

Cobbler 1 
Practically unskilled, including such occupa- 
tions as teamster, expressman, waiter (1), 

day laborer (1), etc. 38 
Total industrial 113 
Proportion of all fathers 20.2% 

Public Service Group: 

Postmen and postoffice clerks 8 

City firemen 4 

Army and Navy officers 3 

Soldiers and sailors 3 

Mayors 2 

Other city officials 3 

Policemen 1 

Civil service clerks 1 
Total public service 25 

Proportion of all fathers 4.5% 

Proportion Proportion in Per cent of 
among fathers population of quota among 

ttiimmnrv of gifted Los A. and San F. fathers of 

ouuunuii/. children (1910 census) gifted children 

Professional Group 29.1% 2.9% 1003% 

Public Service Group 4.5% 3.3% 137% 

Commercial Group 46.2% 36.1% 128% 

Industrial Group 20.2% 57.7% 35% 

In the industrial group only one man gives his occupa- 
tion as "laborer," which is 0.2 per cent of our fathers as 
compared with 15.0 per cent of the general population classi- 
fied as laborers in the census report. Accordingly, fathers 
of gifted children yield only one seventy-seventh of their 
quota for this class. The man referred to was a farmer 
who had moved to Berkeley and taken a position as laborer 
at the University of California in order that his children 
might attend college. 



64 TRAITS OF GIFTED CHILDREN 

Re-grouping according to Taussig's five-grade classifica- 
tions, we have the following : 

1. Professional, including, besides the professions listed 
above, editors and publishers, army and navy officers, may- 
ors and city officials. The total of the professional group 
now becomes 176. 

2. Semi-professional and business group, including two 
sub-grades of "white collar" workers below the professional 
level, (a) Executives and business managers, sales and in- 
surance agents, wholesale dealers, brokers, owners of large 
retail establishments, manufacturers, expert accountants, 
photographers, lithographers, and landscape gardeners. 
Total, 175. (b) Retail dealers and owners of small stores, 
clerical workers, druggists, contractors, florists, telegraph 
operators, postmen and postoffice clerks, civil service clerks. 
Total, 105. Total of 2(a) and 2(b), 280. 

3. Skilled labor group, including carpenters, mechanics, 
machinists, tailors, butchers, farmers, painters, foremen, 
pattern makers, potters, bakers, cobblers, barbers, city fire- 
men, soldiers, sailors, and policemen. Total, 66. 

4. Semi-skilled to slightly skilled laborers, including 
teamsters, expressmen, waiters, etc. Total, 37. 

5. Common laborer. Total, 1. 

The percentage distribution for these five classes is as 
follows : 

Professional 31 A% 

Semi-professional and business 50.0% 

(a) Higher group 31.2% 

(b) Lower group 18.8% 
Skilled labor 11.8% 
Semi-skilled to slightly skilled 6.6% 
Common labor 0.13% 

The above data are in line with the findings of others on 
the social origin of superior ability. Some of the most im- 
portant of these findings are summarized below : 

Cattell, fathers of 885 leading American men of science : 

Professional 43.1% (in general population 3.0%) 

Agriculture 21.2% (" " " 41.1%) 

Mfn and trade 35.7% ( " " " 341%) 



RACIAL AND SOCIAL ORIGIN 65 

Edwin L. Clarke, 666 American men of letters : 
Professional classes 49.2% 

Commercial classes 22.7% 

Agricultural classes 20.9% 

Mechanical, clerical and unskilled 7.2% 

Alphonse de Candolle, European men of science : 

100 Foreign 40 French mem- 
Associates of bers-of both 

the Paris Academy London and Ber- 
of Science lin Academies 

Noble, wealthy, gentlemanly 

classes 41% 28% 

Middle class 52% 47% 

Workers, farmers, etc. 7% 25% 

Havelock Ellis, 1,030 British men and women of genius: 

Upper classes (or "good family") 18.5% 

Church 16.7% 

Law 7.1% 

Medicine 3.6% 

Miscellaneous professions 7.8% 

Army and navy 6.1% 

Officials, clerks, etc. 3.2% 

Commercial 18.8% 

Crafts 9.2% 

Yeomen and farmers 6.0% 

Artisans and unskilled 2.5% 

Perhaps only the last three classes above, including 17.5 per 
cent of the cases, would rank below Class 2 of Taussig. The 
proportion of parents of our gifted below this level is 28.6 
per cent. 

Galton, 107 English men of science : 

Nobility and gentlemen 9 cases 

Army, navy, and civil office 18 " 

Professions 34 " 

Bankers, merchants, manufacturers 43 *' 

Farmers 2 " 

Others 1 " 

The data most nearly comparable with our own are those 
of Cattell and Clarke, but the method by which our gifted 
subjects were selected tends to reduce the numbers for the 



66 TRAITS OF GIFTED CHILDREN 

agricultural class and to increase somewhat those of the pro- 
fessional and commercial classes. Notwithstanding minor 
discrepancies, all of these investigations agree as to the ex- 
istence of a very striking social hierarchy with respect to the 
production of superior individuals. There is one respect in 
which the contribution of the present study is unique. Earlier 
investigations had proved nothing more than that the upper 
social strata are more productive of men and women who 
have succeeded in achieving eminence. It has often been 
argued that this superiority in achievement should be cred- 
ited for the most part to the larger opportunity for achieve- 
ment enjoyed by members of the favored classes. Our data 
show that individuals of the various social classes present 
these same differences in early childhood, a fact which 
strongly suggests that the causal factor lies in original en- 
dowment rather than in environmental influences. 

BARR SCALE RATINGS OF OCCUPATIONAL STATUS 

In order to arrive at a more accurate determination of 
the occupational status of the fathers of our gifted children, 
we have made use of the Barr scale of occupational intelli- 
gence. What we want is a hierarchy of the occupations with 
respect to the relative demands which they make upon intel- 
ligence. The census classification, obviously, is not intended 
to serve this purpose, and although anyone can take the 
census categories and arrange such a hierarchy for himself, 
there is a large probable error in any such individual ar- 
rangement. The error results partly from personal bias and 
partly from insufficient information regarding the actual re- 
quirement of the hundreds of occupations which exist. In 
order to reduce the personal equation in rating the intelli- 
gence value of occupations, Mr. F. E. Barr drew up a list of 
121 representative occupations, each definitely and con- 
cretely described, and had 20 judges rate them on a scale of 
to 100 according to the grade of intelligence which each 
was believed to demand. The ratings were then distributed 
and PJE. values were computed for all the occupations* The 
P.E. values (transmuted) express in the case of each occu- 
pation the number of units of intelligence which, according 
to the composite opinion of these 20 judges, the occupation 
demands for ordinary success. 



RACIAL AND SOCIAL ORIGIN 67 

Scale 

Value Occupation Description 

0.00 Hobo _ 

1.54 Odd Jobs 

2.11 Garbage Collector 

3.38 Circus roustabout Does heavy, rough work about circus. 

3.44 Hostler , .Care of horses in livery, feed and sales stables. 

3.57 R. R. Sec. Hand Replaces ties, etc., under supervision. 

3.62 Day Laborer On street, in shop or factory as roustabout. 

3.99 Track layer ..Does heavy work under supervision. 

4.20 Waterworks man A variety of odd Jobs, all unskilled. 

4.29 Miner Digger and shoveller, etc. 

4.81 Longshoreman Loads and unloads cargoes. 

4.91 Farm laborer Unskilled and usually inefficient. 

4.98 Laundry worker Various kinds of work in laundry (practically 

unskilled) 
5.27 Bar tender 

5.41 Teamster 

5.44 Sawmill worker Heavy work, little skill required. 

5.59 Dairy hand Milking, care of stock under supervision. 

5.81 Drayman..... 

5.87 Deliveryman Delivers groceries, etc., with team or auto. 

6.14 Junkman Collector of junk. 

6.42 Switchman Tending switch in R. R. yards. 

6.66 Smelter worker ..Metal pourers, casting collectors, etc. 

6.27 Tire repairer In general automobile repair shop. 

6.85 Cobbler and shoemaker Repairman in shoe shop. 

6.86 Munition worker Average. 

6.92 Barber Not owner. Has charge of chair. 

6.93 Mov. picture operator Operates machine which projects pictures. 

7.02 Vulcanizer Understands the process of hardening rubber. 

7.05 General repairman Repairs broken articles. Uses wood- working tools. 

7.06 Ship rigger ..Installing cordage system on sailing vessels, 

working under supervision. 

7.17 Telephone operator. 

7.19 Cook In restaurant or small hotel. 

7.23 Streetcar conductor 

7.24 Farm tenants , On small tracts of land. 

7.30 Brakeman - On freight or passenger trains. 

7.33 City fire fighter Handles the ordinary fire-fighting apparatus. 

7.39 R. R. fireman On freight or passenger train. 

7.54 Policeman Average patrolman. 

7.71 Structural steel worker Heavy work demanding some skill. 

7.73 Tel. and Tel. lineman 

7.77 Bricklayer 

7.79 Butcher Not shop owner. Able to make cuts properly. 

7.91 Baker 

8,02 Metal finisher Polishes and lacquers metal fixtures, etc. 

8.04 Plasterer Knowledge of materials used necessary. 

8.08 General painter Paints houses, buildings and various structures. 

8.22 Harness maker 

8.40 Tinsmith...., Makes vessels, utensils, etc., from plated sheet 

metal. 

8.49 Letter carrier 

8.50 Forest ranger ... 

8.58 Stone mason 

8.75 Plumber. Av. trained plumber employee. 

8.89 Gardening, truck farming... Owns and operates small plots. 

8.99 Electric repairman Repairs elec. utensils, devices and machines. 

9.28 Bookbinder .Sets up and binds books of all sorts. 

9.37 Carpenter Knows wood-working tools. Can follow direc- 
tions in various processes of wood construc- 
tion work. 



68 TRAITS OF GIFTED CHILDREN 

Scale 

Value Occupation Description 

9.37 Potter Makes jars, jugs, crockery, earthenware, etc. 

9.54 Tailor - Employee in tailoring shop. 

9.72 Salesman In drygoods, hardware, grocery stores, etc. 

10.11 Telegraph operator In small town. 

10.21 Undertaker In small town. Six mo.-yr. spl. schooling. 

10,26 Station agent In small town. Acts as baggage man, freight agent, 

operator, etc. 

10.26 Mechanical repairman .In shop or factory. Keeps machines in condition. 

10.29 Dairy owner and mgr Small dairy, 50-100 cows. 

10.53 Metal pattern maker 

10.54 Wood pattern maker 

10.54 Lithographer Makes prints from, designs which he puts on stone. 

10.76 Linotype operator 

10.83 Photographer City 1000-5000. A, few months* training, experi- 

ence In studio. 

10.86 Detective Traces clues, etc. Employee of detective bureau. 

10.99 Electrotyper Prepares wood cuts. 

11.17 Traveling salesman Sells drugs, groceries, hardware, drygoods* etc. 

11.34 Clerical work..., Bookkeepers, recorders, abstractors, etc. 

11.35 R. R. Pass. Gondr 

11.51 Store kpr. and owner Small town retail dealer gen. or special store. 

11.74 Foreman ....Small factory, shop, etc. 

11.78 Stenographer. -Writes shorthand and uses typewriter. 

12.02 Librarian In small institution or public library, 

12.06 Nurse and masseur Graduate. 

12.74 Chef ....Employed in large first-class hotels. 

12.84 Editor Small paper, considerable job work. 

12.89 Primary teacher No college training, 2 yrs, special training. 

12.96 Landscape gardener 

13.08 Grammar grade tchr .....Normal graduate expects to make profession 

teaching, 

13.20 Osteopath Training equal to college graduate. 

13.21 Pharmacist In town of from 1000-5000 population. 

13.29 Master mechanic Thorough knowledge in his field of mechanics. 

13.30 Music teacher 2-4 years special training, not college graduate. 

13.31 Manufacturer .. .Employs from 10-50 men. Makes simple nrttclcs. 

13.54 Dentist .....Graduate. 2-5 years experience in small town. 

13.58 Art teacher In high school. Three or 4 years* special training, 

13.71 Surveyor Transit man. City or county surveyor. 

13.31 Train dispatcher Must be mentally alert. 

14.45 Land owner and operator... Very large farms or ranches. 

14.70 Musician Successful player or singer in good company. 

15.05 Secretarial work , Private sec. to high state or national officials, 

15.14 High school teacher Coll. or Normal grad. Not the most progressive. 

15.15 Preacher Minister in town of 1000-5000. ColL graduate. 

15.42 Industrial chemist Thorough knowledge of the chem, of mfg. pro- 

cesses. 

15.43 Mechanical engineer.... Designs and constructs machines and machine 

tools. 

15.71 Teacher in college Degree A.B. or A.M. Not the most progressive. 

15.75 Lawyer In town of moderate size. Income |1000~$5000. 

15.86 Technical engineer Thorough knowledge of the processes of an in- 
dustry. 

16.18 Artist High class painter of portraits, etc* 

16.26 Mining engineer ..Thorough knowledge of mining and extraction of 

metals. 
16.28 Architect. Training equal to college grad. 

16.58 Great wholesale merchant... Business covering one or more states. 

16.59 Consulting engineer In charge of corps of engineers. 

16.64 Educational administrator.. Supt. city 2000-5000 Coll. or Normal graduate* 

16.71 Physician 6-8 yrs. prep, above H. S. Income $5000 and up* 

16.91 Journalist Hih class writer or editor. 



RACIAL AND SOCIAL ORIGIN 69 

Scale 

Value Occupation Description 

17.50 Publisher High class magazine and newspaper or periodi- 
cal, etc. 

17.81 University professor Has A.M. or Ph.D., writes, teaches, and does re- 
search. 

18.06 Great merchant Owns and operates a million dollar business. 

18.14 Musician (Paderewski.) 

18.33 High Nat'l official Cabinet officers, foreign ministers, etc. 

18.85 Writer (Van Dyke.) 

19.45 Research leader Like Binet or Pasteur. 

19.73 Surgeon , (Mayo Bros.) 

20.71 Inventive genius , (Edison type.) 

In the use of the scale it is only necessary to compare the 
occupation which is to be rated, with the occupations whose 
scale values are known, and to assign it the value possessed 
by the scaled occupation which it most nearly matches. In- 
termediate values may be used in rating occupations which 
do not appear in the scale. It has been found that different 
judges agree fairly closely in rating the intellectual demands 
of occupations by this scale. It can not be claimed that the 
Barr Scale values correspond exactly to the facts, but they 
unquestionably approximate the facts more closely than 
would the judgments of any one individual. 

In rating the parents of our gifted children for occupa- 
tional status, additional evidence from other parts of the 
blanks was taken into consideration. For example: if a man 
who had not had college training was reported as an 
"electrical engineer," the rating was changed to that of "elec- 
trician" (or "master mechanic," etc.) according to the infor- 
mation which was available from other sources. The reports 
from the field workers' visits to the homes were often found 
useful in this connection. All statements concerning occupa- 
tions were carefully compared with other known facts before 
being rated, and whenever there was doubt, a lower rather 
than a higher rating was assigned, in order to counterbal- 
ance any unconscious bias which might exist in the mind of 
the assistant who made the ratings. 

The most reliable comparatve data on this head are prob- 
ably the results from the United States census. As the occu- 
pational statistics for the 1920 census were not yet available, 
the 1910 census classification of the adult male population 
of the cities of San Francisco, Los Angeles, and Oakland was 
used. 1 

'See Report of the Thirteenth Census, Vol. IV, pp. 560, 584, 600. 



70 TRAITS OF GIFTED CHILDREN 

The problem was to compute a mean occupation rating 
of the entire adult male population of the above cities on the 
Barr Scale. Each occupation listed for these cities in the 
census report was rated on the Barr Scale, and each rating 
was multiplied by the number reported in the corresponding 
occupational group. The process was simply that of deriving 
a weighted average. Interpolations in rating were resorted to 
freely, and in this case, whenever there was doubt, a higher 
rather than a lower rating was given. This is the reverse of 
the procedure which was followed in rating the parents of 
our gifted, and has the effect of bringing the two groups 
closer together than they would have been found to be had 
all the facts been known. It was thought better to err, if at 
all, on this side. 

In dealing with the census figures, the question arose as 
to what procedure should be followed with the very large 
group of adult males for whom no occupation was reported. 
It seemed fairly certain that, upon the whole, these persons 
would rank below the average of the population in general, 
since a large proportion of them would be day laborers, tem- 
porarily unemployed. Two population ratings were worked 
out: the first (A) based on the assumption that the percent- 
age not reported in the census would have had the same 
distribution of occupational ratings as those reported; the 
second (B) by assigning to the group not reported a constant 
rating of 4.0 P.E., which is above that of a day laborer and 
corresponds closely to that of a railroad track layer. The 
latter is probably a generous figure. 

Table 12 gives the distributions, means, and standard 
deviations for fathers of one gifted child, fathers of more 
than one, total fathers of gifted children, and adult males of 
the general population (the latter by two methods of calcu- 
lation). In connection with columns 1 and 2, it should be 
noted that a majority of the siblings of our gifted group have 
not been tested. If all had been tested it is probable that the 
difference between the two columns ^ould have been 
greater. As it is, the difference is 3.82 times the P.E. of the 
difference. In 34 cases the information regarding the father 
was too indefinite to permit of a Barr rating. Designations as 
indefinite as "teacher," "salesman," or "writer" were thrown 
out. It is quite certain that the omitted cases are at least not 
below the average of total fathers of the gifted group. 



RACIAL AND SOCIAL ORIGIN 



71 



Another fact regarding Table 12 that should be noted 
is the absence of ratings higher than 16 for the general popu- 
lation (last two columns). This is due to the fact that the 
census descriptions of the higher professional occupations 
are less definite than those in the Barr Scale. More definite 
information would have raised the ratings of a certain pro- 
portion of the population cases here rated 16, and possibly 
of some rated 15 or 14, but because of the small numbers 
such change would not have affected to any considerable ex- 
tent the average rating for the general population. 

TABLE 12 

BARR SCALE RATINGS FOR FATHERS OF GIFTED AND FOR MALE ADULTS 
OF THE GENERALITY 





Fathers of 


Fathers of 
more than 


Total 


Adult Males of General 




one 


one gifted 


fathers 


Population 


Rating 


gifted child 


child 


of gifted 


Method A 


Method B 


18-18,9 




2 


2 






17-17,9 


4 


4 


8 






16-16.9 


37 


4 


41 


3005 


3005 


15-15.9 


73 


17 


90 


4075 


4075 


14-14.9 


55 


14 


69 


4296 


4296 


13-13,9 


60 


6 


66 


6210 


6210 


12-12,9 


6 





6 


3866 


3866 


11-11.9 


121 


20 


141 


75150 


75150 


10-10.9 


17 


2 


19 


6089 


6089 


9-9.9 


28 


2 


30 


37204 


37204 


8-8,9 


17 


1 


18 


15235 


15235 


7-7.9 


15 


3 


18 


15167 


15167 


6-6.9 


11 





11 


12602 


12602 


5-5,9 


4 


1 


5 


27379 


27379 


4-4.9 


1 




1 


17907 


87671 


3-3.9 


1 




1 


22334 


22334 


Total 


450 


76 


526 


250519 


320283 


Mean 


12.64 


13,51 


12.77 


8.88 


7.92 


S, D. 


2.77 


2.71 


2.78 


3.24 


3.38 



The mean of 12,77 for total fathers of gifted corresponds 
fairly closely to the Barr rating of a stenographer, librarian 
in a small city, or primary teacher. The mean of 7.92 for the 
general population corresponds to the Barr rating of a plas- 
terer, baker, or metal finisher. 

Grouping the data of Table 12, we have the following 
comparison between total fathers of gifted and adult males 
of the generality (Method B) : 



72 TRAITS OF GIFTED CHILDREN 

Adults of the 
Rating Fathers of Gifted Generality 

15 or above 26.8% 2.2% 

12-15 26.8% 4,5% 

9-12 36.1% 37.0% 

6-9 8.9% 13,4% 

3-6 1.3% 42.9% 

ECONOMIC STATUS 

For 170 families of the gifted Dr. Bronson obtained a 
statement (usually from the mother) of the approximate an- 
nual income. These families are not selected. The question 
was asked in consecutive cases at the time of the medical ex- 
amination and less than half a dozen parents refused to 
answer. As will be seen from Table 13, the economic 
status of a majority of the families is fairly comfortable, but 
there are few cases of wealth. The median income reported 
is $3,333; the mean, $4,705. The median in this case is more 
significant than the mean. Sixty, or 35.3 per cent of the f am- 
ilies, report an income below $2,500, which is probably no 
more than the annual earnings of the average skilled laborer 
in California in 1923. Only twenty-nine (17 per cent) report 
the income above $7,500, and seven (4.1 per cent) above 
$12,000. A few families in this group of 170 and several of 
our families not in this group are living in what might truly 
be called poverty. 

TABLE 13 
INCOME REPORTED BY 170 FAMILIES OF GIFTED CHILDREN 



Families 


Income 


7 


$12,500 to 125,000 


4 


11 


,500 


" 12,500 


3 


10 


,500 


" 11,500 


8 


9 


,500 


" 10,500 


2 


8 


,500 


9,500 


5 


7 


,500 


S,500 


5 


6 


,500 


7,500 


12 


5,500 


6,500 


15 


4,500 


5,00 


19 


3,500 


4,500 


30 


2,500 


' 3,500 


52 


1,500 


' 2,500 


8 




500 


1,500 


Median 


$3,333 


Mean 


4,705 


S. D. 


3,805 



RACIAL AND SOCIAL ORIGIN 73 

HOME RATINGS ON THE WHITTIER SCALE 

One-third of the homes of our gifted children were rated 
by the field assistants on the Whittier Scale for Grading 
Home Conditions, devised by Dr. J. Harold Williams. 1 
Observations were made during conferences with parents 
at their homes and the pertinent facts were recorded 
immediately after departure. The 288 homes graded are be- 
lieved to be fairly representative of the group. The scale 
contains a score card with directions for grading each of five 
different items on a scale of 5 to 1 (6 to in exceptional 
cases) : necessities, neatness, size, parental conditions, and 
parental supervision. The meaning of the diff erent grades is 
defined in a very concrete way on the score card for each of 
the items graded. The sum of the separate ratings is the 
home index. Table 14 gives the distribution of Whittier 
scores for each item, with mean and standard deviation. 

TABLE 14 

WHITTIER SCALE RATINGS OF 288 HOMES OF GIFTED CHILDREN 
Distribution of Ratings 

Mean Standard 



Items 





1 


2 


3 


4 


5 


6 Rating 


Deviation 


Necessities 








3 


26 


60 


184 


15 4.63 


.76 


Neatness 





1 


5 


26 


65 


188 


3 4.54 


.77 


Size 





1 


8 


25 


60 


185 


9 4.55 


.83 


Parental Conditions 


1 





6 


24 


41 


208 


? 4.64 


.81 


Parental Supervision 





1 


4 


27 


74 


153 


29 4.60 


.81 


Total score, or index 














22.94 


3.00 



Below are given the means in comparison with the means 
reported by Dr. Williams for 50 unselected homes and for 
120 homes of delinquent boys. 





Unselected 


Homes of 


Homes of 




Homes 


Delinquents 


Gifted 




(50) 


(120) 


(288) 


Necessities 


4.18 


2.93 


4.63 


Neatness 


4,20 


3.39 


4.54 


Size 


4.48 


3.11 


4,55 


Parental Conditions 


4.22 


2.64 


4.64 


Parental Supervision 


3.70 


1.84 


4.60 



Total Score 20.78 13.91 22.94 

'Described in The Journal of Delinquency, Vol. 1, 1916, pp. 271-286. 



74 TRAITS OF GIFTED CHILDREN 

In the case of the gifted, the ratings for the separate items 
tend to be very uniform, with a low S.D. for each. On all 
items the homes of the gifted rate much higher than those of 
delinquents, with the greatest difference in parental super- 
vision. The unselected homes approximate those of the 
gifted on all items except parental supervision. The differ- 
ence in mean home index of 22.9 for gifted and 13.9 for 
delinquents is very large. 

The total ratings of the 288 homes of gifted children were 
distributed as follows : 

Total Rating 6-8 9-11 12-14 15-17 18-20 21-23 24-26 27-29 
Cases 11 4 7 41 86 137 11 



DIVORCE AND SEPARATION OF PARENTS 

Data on this point are of interest not only in connection 
with the general quality of home environment our gifted 
children have enjoyed, but also as an indication of the fre- 
quency with which the parents of such children are socially 
unadaptable or temperamentally exceptional. The Home 
Blank contained the following questions : 

Are parents divorced? . ... If so, when? . . , . 
// not divorced, are parents separated? . . . How long 
separated? .... 

The data thus obtained were in many cases checked up at 
the time of home visits. The facts as reported are as follows 
for the 578 families of our main group : 

N Per cent 

Divorces reported or otherwise known 30 5.2% 

Parents separated 11 1.9% 

Facts not known 1 5 .8% 

Families known to be unbroken 532 92.1% 

There are no strictly comparable data for the general 
population from which these parents come. In 1916 there 
were 5,573 divorces and 30,996 marriages in California, or a 

a ln three of the five cases the gifted child had been adopted in In- 
fancy; information regarding true parents was not obtainable, but the 
foster parents were living together. 



RACIAL AND SOCIAL ORIGIN 75 

ratio of 1 to 5.56. The ratios were as follows in the three 
counties with which we are here chiefly concerned: 

Los Angeles County, 1 to 5.24 

San Francisco County, 1 to 4.35 

Alameda County, 1 to 4.71 

The three counties combined, 1 to 4.77 

However, the proportion is lower for fruitful than for barren 
marriages, and we are here concerned with fruitful mar- 
riages only. In regard to the 5,573 divorces granted in Cali- 
fornia in 1916, certain additional facts were available in 5,196 
cases. In the case of 2,088 of these 5,196 divorces, dependent 
children were reported. This is 40.2 per cent. The report was 
"no dependent children" in the case of 57.6 per cent, while in 
2.2 per cent this item was not reported. Roughly speaking, 
one may say that for 100 marriages in California there are 
18 divorces and that in 7 (or 40 per cent) of these cases there 
are dependent children. Assuming that 80 of the 100 mar- 
riages are fruitful, we may take the ratio of 7 to 80 (or 8.75 
per cent) as the figure with which to compare the 5.2 per cent 
of divorces in the families of our gifted. Making considerable 
allowance for various sources of error in this comparison, it 
seems reasonably certain that the divorce records for our 
578 families are better than for the population to which they 
belong. 

NEIGHBORHOOD RATINGS 

Dr. Williams has devised a scale for rating neighbor- 
hoods similar to that for rating homes. 1 It is unfortunate that 
lack of time prevented the use of this scale for rating the 
neighborhood environment of our gifted. Instead, a simpler 
method was used. In visiting a home the field assistant ob- 
served the surroundings and recorded an unanalyzed rating 
of 1 to 5, as follows : 

1 Very superior 

2 Superior 

3 Average 

4 Inferior 

5 Very inferior 

*A Scale for Grading Neighborhood Conditions, Bulletin No. 5, Depart- 
ment of Research, Whittier State School, Whittier, California, 1917, p. 17. 



76 TRAITS OF GIFTED CHILDREN 

The average rating of 305 neighborhoods, chiefly those in 
which the homes that had been rated were located, was 2.25, 
S.D. 0.94. The distribution was as follows : 

Very superior 76 

Superior 108 

Average 92 

Inferior 28 
Very inferior 

No comparative data for unselected homes are available, 
but if the ratings have been given as intended by the scale, 
the average neighborhood in which the gifted child resides is 
about the same as the average for the cities in question. We 
have seen that the average home index, however, is very 
superior, notwithstanding the modest average income. The 
conclusion seems to be that although the financial status of 
the parents of the average gifted child is such as to necessi- 
tate living in an average neighborhood, the intelligence and 
character of the parents are such as to insure that the inter- 
nal conditions of the home will be above the average. The 
data presented later in this chapter on the education of par- 
ents support this conclusion. 

SCHOOL REPORTS ON HOME ENVIRONMENT 

In the School Blank the teacher was asked to give signifi- 
cant facts regarding the child's home environment (e.g., im- 
perfect parental control, excessive indulgence, undue sever- 
ity* systematic home instruction, unsuitable companions, 
etc.). The question was answered for 417 gifted children 
and for 319 of the control group (unselected children). 1 The 
results were tabulated separately by sex, but have been here 
combined, as no significant sex differences were found* It 
should be noted that the wording of the question is such as 
hardly to call for a reply unless "significant facts" regarding 
the child's home environment were known to the teacher; 
also that the question is especially designed to bring to light 
unfavorable conditions. Probably in a large majority of 
cases failure to answer the question means that no especially 
significant facts, particularly facts of an unfavorable nature, 

a See pp. 177-178 for description of this control grotip. 



RACIAL AND SOCIAL ORIGIN 77 

were known to the teacher. However, the percentages have 
been based on the number for whom the question was an- 
swered. Probably the most significant fact is that unfavor- 
able home conditions were mentioned for only 8.6 per cent 
of the gifted as compared with 24.1 per cent for the control 
group. 

TABLE 15 
SCHOOL REPORTS ON HOME ENVIRONMENT 

Gifted Control 

N Per cent N Per cent 

Replies received 417 319 

A. Probably favorable circumstances 355 85.1% 197 61.8% 

1. Systematic home instruction 181 43.4% 98 30.7% 

2. "Good environment" 171 41.0% 97 30.4% 

3. Well educated parents 3 0.7% 0.0% 

4. Travel, excellent associations 0.0% 1 0.3% 

5. Excellent companions 0.0% 1 0.3% 

B. Probably unfavorable circumstances 36 8.6% 77 24.1% 
1. Excessive indulgence 14 3.4% 14 4.4% 

*2. One or both parents dead 5 1.2% 14 4.4% 

*3. Parents divorced 2 0.5% 2 0.6% 

4. "Poor home conditions" 2 0.5% 8 2.5% 

5. Imperfect parental control 6 1.4% 12 3.8% 

6. Foreign home 3 0.7% 10 3.1% 

7. Unsuitable companions 0.0% 2 0.6% 

8. Undue seventy 1 0.2% 9 2.8% 

9. Child obliged to care for home 0.0% 2 0.6% 

10. Has to look after himself 1 0.2% 3 0.9% 

11. Left to care of nurse 2 0.5% 0.0% 

12. Family below average mentality 0*0% 1 0.3% 

C. Unclassified 26 6.2% 45 14,1% 

1. "Average home" 14 3.4% 31 9.7% 

2, Miscellaneous minor comments 12 2.8% 14 4.4% 

* These two items arc incomplete. Probably these represent cases which have 
attracted teachers* attention because of unfortunate associated circumstances or 
results* 

PAID EMPLOYMENT OF GIFTED CHILDREN 

In the Home Blank suitable spaces were provided for re- 
porting the paid employment the child had had, together 
with kind, age when begun, number of hours weekly, and 
length of time continued. 

Of 330 boys, 52, or 16 per cent, have at some time had paid 
employment outside the home. In addition to this, another 



78 TRAITS OF GIFTED CHILDREN 

16 per cent have received payment for definite services 
within the home or to immediate neighbors. 

Of 273 girls, 5, or less than 2 per cent, have been employed 
outside the home, and 33 have been paid for services within 
the home. 

Salaries range from trifling sums up to |400 a week, the 
latter for motion picture stars, 1 

Ages at the time of first employment range from two 
years (in motion pictures) up to twelve. 

Kinds of employment (outside of home) : 

No. Cases 

1. Paper routes 30 

2. Delivery boy, errand boy 22 

3. Motion pictures or stage work 7 

4. Clerk in store * 

5. Church choir ^ 

6. Miscellaneous minor employment 21 

Total . 77 

More th&n one kind of employment listed for same child 20 

Total cases 57 

EDUCATION OF PABENTS AND GRANDPARENTS 

The data for parents consist of reports on the following 
item in the Home Blank: Draw a line under the highest 
school grade reached: Father: i f 2, 3, 4, 5, , 7, 8; High School 
i, 2, 3, k; College 1, 2, 3, k; Post Grad. Other kinds of schools 
attended by father, and length of time . * , . , The question 
was asked separately for mothers. For each grandparent the 
question asked was simply: Extent of education (school 
grade reached). . . . 

The figures of Table 16 may be too low, owing to a nat- 
ural tendency to interpret the term "grade reached" ^as 
meaning grade completed. It will be noted that the median 
grade reached by the parents was 12.1 (which falls Just 
within the first year of college) and that there was little 
difference between the amount of schooling of fathers and 
mothers. 

a The most famous of child movie stars Is known to have tested above 
140 IQ, but we failed to secure the cooperation of his parents to the extent 
of having him included in our gifted group. 



RACIAL AND SOCIAL ORIGIN 



79 



*& t3 

I S 











> 


!>. 


tO 


04 


TH 


^ 


r* 


TH 








* 


^ 


04 


04 


04 


04 

TH 


04 

TH 


O4 


10 

TH 






bO 


co 


CO 


04 

TH 


CO 

TH 


TH 


O4 


^ 


^f 


OS 






"o 
























o 


0* 





00 


!> 


00 


CO 


CO 


*f 


l-v. 










TH 




TH 


TH 


to 


O4 


TH 


CO 










Jt> 


^ 


uO 


04 


oo 


O4 


Tf* 


CO 




2- 




Y-< 






TH 


TH 


CO 


O4 


TH 


CO 




55 






r-. 


O 


CO 


CO 


oa 


t^ 


04 


OS 




3 


_ 


^ 


10 




00 




04 


C 


CO 

TH 


04 
04 




S 


o 

rl 






















^ 


1 


oo 


04 


04 

TH 


TH 

04 


TH 


r- 





CO 


CO 




p. 
























1 




CM 


O 
04 


O4 
04 


10 
O4 


04 


00 


TH 


CO 


00 




a 






























04 


10 


CO 


CO 


CO 


CO 


^4 


|><* 




Pt4 




T"* 


O4 


TH 




TH 




04 


1O 


! 




O 
























t/3 






TH 


O4 


CO 


OS 


oo 


CD 


o* 


OS 




E"* 




00 


*o 


1O 






Oi 


CO 


04 


TH 



















TH 


04 


CO 


CO 


CO 


1 


. 


^ 


00 


0> 


oo 


CO 


00 
04 


10 

TH 


TH 


04 

CO 


TH 


O 























S 


1 


1 


CD 


03 


O4 

TH 


TH 


CO 


TH 

CO 


04 
04 


CO 
O4 


10 


< 





^ 




















H 




3 


iO 


"* 


04 


04 


TH 


05 


TH 


TH 


TH 

04 




55 


a 






















"^ 


1 




TH 


CO 


TH 


04 


l> 


CO 


TH 


Tt< 




13 


3 














TH 


TH 


,04 




S 




C* 


TH 


O4 


I 


O4 


10 


TH 


CO 


- 




a, 




O4 


I 


CO 


I 


**< 


t- 


U3 


TH 


CO 




o 




TH 


1 


I 


I 


I 


1 


CO 


04 


10 




CHOOLING 


-2 


Schooling*' 





' 


04 


1 


r- 


0, 


TH 

TH 


o 

04 




CO 






























S. 


-2 





CO 

T: 


















r 


S 


S 


"o 
















Parents 


Fathers of Gifted 


Fathers of Gifted 


Mothers of Gifted 


Mothers of Gifted 


All Parents 


Grandparents 
All Grandfathers 


All Grandmother* 


All Grandparents 



80 TRAITS OF GIFTED CHILDREN 

The central tendencies and variabilities for the data of 
Table 16 are as follows: Amount5 of Schooling !n Grades 

Median Mean S, D. 

Fathers of Gifted Boys 12.2 12.6 3*99 

Girls 11.3 11.8 3.74 

Mothers " " Boys 12.2 11.9 3.10 

Girls 12.2 11.9 3.36 

All Parents 12.1 11.8 3.46 

All Grandfathers 8.9 10.8 4.05 

All Grandmothers 8.8 9.7 2.84 

All Grandparents 8.9 10.0 3.42 

Parents had had schooling in addition to that shown in 
Table 16 as follows: 

Business School 114 

Normal School 48 

Law School (not included under college) 26 

Evening School 21 

Medical School (not included under college) 18 

Conservatory of Music 16 

Art School 10 

Dramatic School 9 

Technical School 8 

Nurses' Training School 7 

Travel and Study abroad 7 

Miscellaneous 49 

Total ' ' ~ ' "~~~ "~333" 

Per cent of total group 27,5% 

Reports are available for only about two-thirds of the 
grandparents. Table 16 shows the median grade reached 
by the grandfathers and grandmothers to be 8.9, or nearly 
through the ninth grade. However, the mean for the grand- 
fathers is 10.8 as compared with 9.7 for the grandmothers. If 
the schooling of all grandparents had been reported, the 
means and medians would probably have been lower. 

The Home Blank provided separate spaces for reporting 
the college degrees each parent had obtained and where 
they were obtained. In treating the results on this item only 
colleges and universities of recognized standing were in- 
cluded. All degrees requiring less than four years of study 
after high school graduation were excluded (as DJD-S.; also 
D.D., when it was obtained from a theological seminary of 
less than standard college grade). For this reason, the per 
cents given below are somewhat lower than they would have 
been had they been based on the data of Table 16, 



RACIAL AND SOCIAL ORIGIN 81 

Number of Per cent of 
children all children 

Father, only, holds degree 98 16.2% 

Mother, only, holds degree 18 3.0% 

Both parents hold degree 44 7.3% 

One or both parents hold degree 160 26.4% 

That is, more than a quarter of these children have at least 
one parent who is a college graduate. There were 204 indi- 
vidual parents who were college graduates, which is 16.9 per 
cent of the entire number. There are no comparative data 
on the proportion of adults of corresponding age in the gen- 
eral population who are college graduates, but it is doubtful 
whether it would be more than one-fifteenth or one-twentieth 
of the proportion found for this group. 

The median amount of schooling for the native born 
white draft of the United States army in the recent war, as 
given in the official report 1 was 6.9, or not quite through the 
seventh grade. For the same group the proportion of college 
graduates was approximately 1 per cent The figures re- 
ported for the army agree closely with those of Thorndike 
based upon school reports. 3 In neither case, however, are 
the data exactly comparable with those for parents of our 
gifted. These should be compared with adults in the general- 
ity of corresponding age. The mean age of the United States 
army draft of 1917 and 1918 was probably fifteen years lower 
than that of the parents of our gifted, and as the mean 
amount of schooling in the general population has risen con- 
siderably in recent years, the showing made by our parents 
is even better than it at first appears. 

SIZE OF HOME LIBRABY 

One indication of the cultural status of a home is the 
number of books it contains. The Home Blank contained the 
following item : Jot down a rough estimate of the number of 

books in the home library This information was sought 

primarily for its significance in connection with individual 
children, but the figures are also of interest for the group. 
Unfortunately, no comparative data are at hand for unse- 



'Psychological Examining in the U. S. Army. Memoirs of the National 
zdemy of Science, Vol. I 
a E. L- Thorndike: The ! 
of Educ, BulL, 1907, No. 4. 



Academy of Science, Vol. 15, p. 761. 

a E. L- Thorndike: The Elimination of Pupils from School. U. S. Bur. 



82 TRAITS OF GIFTED CHILDREN 

lected children, but such would probably show but a small 
fraction of the mean (328) reported for our gifted. The re- 
sults for the 547 children for whom reports are available are 
as follows: 



No. I**T cent 



"No books" 7 


1,1% 


10 or fewer 15 


2.3% 


25 




41 


6.4% 


50 




119 


18.5% 


100 




215 


33.4% 


500 o 


r m 


:>re 106 


16.5% 


750 




57 


8.9% 


1000 




43 


6.7% 


2000 




6 


0.9% 



Total range reported, to 6000 
Median 202 

Mean 328 

S. D. 458 

It is impossible to say how seriously the above reports 
have been affected by constant errors in the direction of 
overestimate or underestimate. It is unlikely that serious 
intentional exaggeration occurred in many cases, but it is 
conceivable that honest estimates of this kind tend to run 
either too high or too low. 

SUMMARY 

1. Data on racial origin indicate that, in comparison with 
the general population of the cities concerned, our gifted 
show a 100 per cent excess of Jewish blood; a 26 per cent 
excess of parents who are of native parentage; a probable 
excess of Scotch ancestry; and a very great deficiency of 
Latin and negro ancestry. 

2. Half of the parents were born in cities of 10,000 popu- 
lation or over, and almost a quarter in cities or towns of 
1,000 to 10,000, leaving only a quarter for rural districts and 
towns or villages of less than 1,000. The grandparents were 
only slightly of tener of rural origin than were the parents. 

3. Classification of the occupations of fathers into the five 
grades of Taussig gave 31.4 per cent for Class I (profes- 
sional) ; 50 per cent for Class II (semi-professional and 
business); 11.8 per cent for Class III (skilled labor); and a 



RACIAL AND SOCIAL ORIGIN 83 

total of 6.8 per cent for Classes IV and V (semi-skilled and 
unskilled labor). Earlier investigations have shown that 
social class is highly correlated with adult achievement; this 
study shows that it is highly correlated with intelligence in 
fairly early childhood. 

4. Ratings of the occupations of parents on the Barr 
Scale give a mean which is very far above the mean Barr 
rating for the general population. 

5. The median income of 170 random homes of the gifted 
was |3,333, with mean of $4,705 and S.D. of $3,805. 

6. Ratings of 288 random homes of the gifted on the 
Whittier Scale for home grading yielded a mean score above 
that for unselected homes, and far above that for homes of 
delinquent boys. 

7. In the case of 5.24 per cent of families of the gifted, the 
parents are divorced, and in the case of 1.9 per cent they are 
separated. These figures are lower than those for fruitful 
marriages in the general population of California. 

8. The neighborhoods in which 305 random homes of 
gifted children were located gave a mean rating of only a 
little above "average." 

9. The school reports unfavorable home conditions for 
only 8.6 per cent of the gifted group, as compared with 24.1 
per cent of the control group. 

10. Approximately 16 per cent of our boys and 2 per cent 
of our girls have at some time had paid employment outside 
the home. The maximum salary was $400 a week. 

11. The mean amount of schooling for both fathers and 
mothers is approximately 12 grades, with SJD. of about 3.5 
grades. The mean for grandfathers is 10.8, and for grand- 
mothers, 9.7, The average parent of the gifted child has cov- 
ered about twice as many school grades as the average adult 
in the population. 

12. A quarter of our subjects have at least one parent 
who is a college graduate, and about 17 per cent of the par- 
ents hold a degree from a college of standard grade. 

13. Libraries in the homes of our subjects range in size 
from to 6,000 books, with median of 202 and mean of 328 
(S.D. 458), 

14. The data of this chapter offer considerable indirect 
evidence that the heredity of our gifted subjects is much 
superior to that of the average individual. 



CHAPTER V 



INTELLECTUALLY SUPERIOR RELATIVES 1 

It has thus far not been possible to undertake an ex- 
tended study of the heredity of our gifted children, but the 
facts presented in the preceding chapter indicate that a ma- 
jority have sprung from families of distinctly better than 
average ability. The purpose of this chapter is to summarize 
such information as we have been able to secure with refer- 
ence to the frequency of gifted relatives. Four pages (8% xll 
inches) of the Home Information Blank were devoted to 
questions regarding family history. 2 The questions follow : 

13. LIVING BROTHERS AND SISTERS of this child: (Mark with cross (X) any half- 
brothers or sisters) 

First name Boy or Present School grade Intelligence of each com- Occupation if 
girl age reached pared with this child not in school 



2. 

3T 

~ 

~ 
"67 



14. DECEASED BROTHERS AND SISTERS: (Mark with cross any half-brothers or sisters) 

Boy or Age at School grade Intelligence of each as 

girl death reached Cause of death compared with this child 



1, 










2. 










3. 










4. 











15. Is child's FATHER living or dead? Age at death Cause.. 

16. Is child's MOTHER living or dead? Age at death Cause.. 

17. Are parents divorced? Jf so, when? 

18. If not divorced, are parents separated? ...How long? 



V. FATHER'S RELATIVES. 

1, FATHER'S FATHER: Name... Birthplace 

(a) Age If living If dead, give age at death Cause of death 

<b) Descent (for example, English and Scotch, German, Dutch and French, 
Russian Jewish, etc.) 

'Written with the assistance of Florence L. Goodenough. 
*A part of the information thus collected has been presented in other 
chapters. 

85 



86 



TRAITS OF GIFTED CHILDREN 



(c) His occupation or occupations 

(d) Extent of his education (school grade reached) .. 

(e) As compared with the strictly average person, was his intelligence very 
superior, superior, average, inferior, very inferior? (Underline) 

Add any other information you can that would help to give an idea of his 
intelligence 



(f) His special interests, hobbies, or accomplishments.. 



(g) Positions of honor, trust, or recognition.. 



(h) His outstanding characteristics.... 



2. FATHER'S MOTHER: Maiden name Birthplace ,. 

(a) Age if living If dead, give age at death ,. .. Cause of death, 

(b) Of what descent? , .. 

(c) Her occupation or occupations ,. .....>.,.... ... ... 

(d) Extent of her education (school grade reached) , 

(e) As compared with the strictly average person, was her intelligence very 
superior, superior, average* inferior, very inferior? {Underline) 

Add any other information you can that would help to give an idea of her 
intelligence 



(f) Her special interests, hobbies, or accomplishments... 



(g) Positions of honor, trust, or recognition.. 



(h) Her outstanding characteristics.. 



3. FATHER'S BROTHERS AND SISTERS: (Mark with cross any half-brothers or half- 
sisters) 

Intelligence vry 

Present agd superior, superior. 

Living (or age at average, inferior. 

Name Occupation or dead death) Cause of death or very inferior? 



INTELLECTUALLY SUPERIOR RELATIVES 87 

4. Relatives more distant (on FATHER'S side) of exceptional ability. Give relation- 
ship to FATHER, and state accomplishments 



5, What occupations have been most common on FATHER'S side?.. 



6. Please answer this item fully regarding relatives on FATHER'S side. (Information 

is confidential.) 

Were any insane? Number Relationship to father 

Were any decidedly "queer**? Number Relationship to father. 

Were any epileptic? Number Relationship to father 

Were any mentally backward? Number Relationship to father 

Did any commit suicide? Number Relationship to father 

7. Other information regarding FATHER'S relatives 



VI. MOTHER'S RELATIVES. 
1. MOTHER'S FATHER: Name Birthplace 

(a) Age, if living If dead, give age at death Cause of death...- 

(b) Of what descent? 

(c) His occupation or occupations 

(d) Extent of his education (school grade reached) 

(e) As compared with the strictly average person, was his intelligence very 
superior, superior, average, inferior, very inferior? (Underline) 

Add any other information you can that would help to give an idea of his 
Intelligence 



(f) His special interests, hobbies, or accomplishments.. 

(g) Positions of honor, trust, or recognition , 

(h) His outstanding characteristics 



2. MOTHER'S MOTHER: Maiden name Birthplace 

(a) Age, if living..... If dead, give age at death Cause of death 

(b) Of what descent - 

(c) Her occupation or occupations ...- - 

(d) Extent of her education (school grade reached) - 

(e) As compared with the strictly average person, was her intelligence very 
superior, superior, average, inferior, very inferior? (Underline) 

Add any other information you can that would help to give an idea of her 
intelligence 



8$ TRAITS OF GIFTED CHILDREN 

(f) Her special interests, hobbies, or accomplishments.. ..... 



(g) Positions of honor, trust, or recognition. 



(h) Her outstanding characteristics. 



3. MOTHER'S BROTHERS AND SISTERS: (Mark with cross any half-brothers or half- 
sisters) 

Intelligence very 

Present age superior, superior. 

Living (or age at average, inferior. 

Name Occupation or dead death) Cause of death or very inferior? 



1. 



2. 
_ 

~ 
57 
6T 

~ 



4. Relatives more distant (on MOTHER'S side) of exceptional ability. Give rela- 
tionship to MOTHER and state accomplishments...,..,. .,.,.,.....,.,..... 

5. What occupations have been most common on MOTHER'S side?, ......,,....... 

6. Please answer this item fully regarding relatives on MOTHER'S side. (Informa- 
tion confidential.) 

Were any insane? Number. Relationship to mother.,,,,.. .... 

Were any decidedly "queer"?,. Number.,., .......Relationship to mother...,,,, ., 

Were any epileptic? , ..Number ...Helationship to xnothfr..,..,, , . 

Were any mentally backward?, Number,,, ...... ..Relationship to mother,,., ,.,, , 

Bid any commit suicide?.. .........Number.. .Relationship to mother.,,.,,..* . 

1. Other information regarding MOTHER'S relatives ,,..,,,...,..., .,..,......,,..,.,. 



ON THE REMAINING PAGES GIVE ANY ADDITIONAL INFORMATION 
REGARDING THE CHILD WHICH YOU THINK WOULB BB OF INTEREST; 
FOR EXAMPLE, INDICATIONS OF SUPERIOR INTELLIGENCE, INFORWA- 
TION REGARDING EARLY MENTAL DEVELOPMENT, RELATIONS WITH 
OTHER CHILDREN, RELATIVES OF SUPERIOR ABILITY, CHILD'S OUT- 
STANDING CHARACTERISTICS, ETC. 



INTELLECTUALLY SUPERIOR RELATIVES 89 

The Home Information Blank was filled out for 94 per 
cent of the children in our main group, but not always with 
the completeness that could have been desired. 1 However, 
there were few questions which were not answered in as 
many as 90 per cent of the blanks returned. The trouble lay 
not so much in failure to answer, as in the failure to record 
all the facts that would have been significant. Sometimes 
the facts were not known, but frequently they were omitted 
simply because the parents did not consider them important. 
The standards of performance and culture in these families 
are so high that a relative who is a highly successful phy- 
sician or lawyer or banker is likely to be described as of 
"average ability." Success is taken for granted. There were 
many cases of failure to mention fairly close relatives who 
were really distinguished. In the case of one subject, for 
whom seven eminent relatives were listed in the Home 
Blank, further investigation brought to light 34 of sufficient 
distinction to be included in Appleton's Cyclopedia of Amer- 
ican Biography or Who's Who! Doubtless in many cases the 
mere labor of writing down the information requested was 
the deterring factor. The small amount of space provided in 
the Home Blank in which to record facts about gifted rela- 
tives also limited the information secured. Home visits by 
the field assistants and correspondence have supplemented 
considerably the data supplied in the Home Blank, but our 
information on heredity is still extremely fragmentary. 

SIBLINGS IN THE MAIN GROUP 

The 643 children discovered in the main search belong to 
578 families. At the time the search was conducted 8 other 
children in these families qualified on the Terman Group 
Test for the high school gifted group. One qualified in the 
outside Binet group, and 5 were included because they came 
within five points of qualifying on the Binet. This brought 
the number to 657, or an average of 1.136 per family. Not 
counting the 5 cases with slightly below standard IQ, the 
total is 651, or an average of 1.126 per family. A year later 
19 additional cases in these families qualified as a result of 
testing 48 sibs who had not been tested before. At present, 

'Blanks for several additional families were secured after this sum- 
mary was compiled. 



90 TRAITS OF GIFTED CHILDREN 

therefore, the 578 families have furnished 676 subjects, or an 
average of 1.17 per family. There are at least 481 sibs (the 
number is growing) who have not been tested, and we esti- 
mate that among these are probably between 50 and 75 who 
are capable of reaching the standard set for the group. Be- 
low, will be found the number* of families with one, two, 
three, four, or five children who qualified for the gifted 
group. Two sets of figures are given : A, those based upon 
the 658 children discovered in the original search (643 of 
main group and 15 others) ; and B, those which include in 
addition the 18 cases discovered by tests of sibs a year later. 

A B 

Families furnishing one gifted subject 511 497 

two " subjects 62 73 

three " 24 

four " 34 

five " II 

_____ 

The cumulative figures are as follows: 

A B 

Families furn shing five gifted subjects 1 I 

four or more 4 5 

three or more 6 9 

two or more 68 80 

one or more 578 578 

The number of families furnishing more than one sub- 
ject would, of course, have been much smaller had no special 
effort been made to test sibs who had already qualified in the 
usual way as a result of the school search. We estimate, 
however, that even apart from this advantage, 578 families 
would have yielded approximately 600 subjects. Considering 
the highly selected nature of our group, the likelihood of a 
family furnishing two subjects by mere chance is very re- 
mote. Not more than one child in two hundred in the general 
school population is capable of satisfying the standard. The 
families represented in our group probably have on an aver- 
age not more than two children attending the public schools. 
Reckoning the number in school as two, a given family 
would have one chance in a hundred of furnishing one sub- 
ject. The likelihood of one family furnishing two subjects by 

chance alone would be nA^ or one in ten thousand, Ac- 



INTELLECTUALLY SUPERIOR RELATIVES 



91 



cordingly, by chance alone we should not have expected a 
single family to have yielded two subjects. Actually, 71 fur- 
nished two, and 80 two or more. Taking our estimate that 20 
families would have done so even if no special search had 
been carried on among sibs, this would be 346 times the 
number which chance alone would have given. The record 
of 71 families with two children is, on the same basis of reck- 
oning, 1,228 times as high as chance would give. In so far as 
gifted sibs have been missed, this record is lower than it de- 
serves to be. On the other hand it is possible that teachers in 
nominating children for the tests may have been inclined to 
favor the sib of a child who had made an exceptional school 
record. However, even if we make considerable allowance 
for this possibility, the showing made by these 578 families 
must still be considered remarkable. 

RELATIVES IN THE HALL OF FAME 

Of the 62 members of the Hall of Fame, 14, or 22.58 per 
cent, are known to be related to one or more children of our 
main gifted group. At least one Hall of Fame relative was 
reported for 15 children, or 2.3 per cent of the group; and 
two or more for 7 children. The relationships are given in 
Table 17. 



Name 

1. John Adams 

2. John Quincy Adams 

3. Henry Ward Beecher 

4. Samuel L. Clemens 



TABLE 17 
RELATIVES IN THE HALL OF FAME 

Relationship 
Direct line 
Direct line 
Not stated 
"Distant cousin" 



5. Charlotte Cushman 

6. Benjamin Franklin 

7. U. S, Grant 

8. Elias Howe 

9. Andrew Jackson 
10* H. W. Longfellow 



"Third cousin" 
Not stated 

"Gt.-gt.-gt.-gt.-great uncle" 
"Third cousin, once removed" 
Direct line 

Not stated 
Gt.-great-uncle 



11, Harriet Beecher Stowe Not stated 



Child 

E.A. 

E.A. 

J. W. S. 

M.G. 

(L.K. 

)R.K. 

R.N, 

(R.K. 

)R.K. 

P.R. 

(L.K. 

|R,K. 

Norma T. 

P.R. 

J. W. S. 



92 TRAITS OF GIFTED CHILDREN 

TABLE 17 Concluded 

Name Relationship Child 

12. George Washington (In direct line from Ailing (R.E. 

and Mary Ball, grandparents |M. E. 
of George Washington.) 

13. J, Greenleaf Whlttier Not stated A. M. 

[Direct line V. M. 

14. Roger Williams Direct fine ft^ 

[Not stated ) R ; 

RELATIVES IN WHO'S WHO 

Twelve of the 643 are known to have a parent or grand- 
parent in Who's Who (1921-22). These 12 children represent 
8 families, which gives an average of L5 gifted children for 
such families. There are three fathers, two mothers, and four 
grandfathers in Who's Who. Two sibs have mother and 
grandfather thus distinguished. Each of the two mothers 
has two children in the group (all the living offspring). Of 
the three fathers, one has two children in the group. All the 
children of these three fathers who have been tested have 
qualified. 

The above numbers are almost certainly too small, as it 
has not been possible to check all parents and grandparents 
against the numerous editions of Who's Who. It is also 
necessary to take into account the fact that half or more of 
these parents are still below the average age of first inclusion 
in Who's Who. Other parents of the group will doubtless yet 
attain this degree of distinction. The Who's Who group 
would have been much larger had the faculties of Stanford 
and the University of California been canvassed. Stanford 
does not enter at all in what we have called the "main 
search," and the University of California faculty was only 
partly canvassed. Of 128 old cases whom we have been fol- 
lowing for several years, 12 are known to have a parent in 
Who's Who. 

That three of the 578 fathers and two of the 578 mothers 
in our main group should have attained this distinction by 
mid-life is many times the number chance would give. Brim- 
hall 1 finds the average age of first inclusion in Who's Who 
(Vol. 6) to be 49.9 years. The average age of our 578 fathers 

*Dean R. Brimhall: Family Resemblances Among American Men of 
Science. The American Naturalist, 1923, p. 81, 



INTELLECTUALLY SUPERIOR RELATIVES 93 

is about 41 years, and that of our 578 mothers about 37 years. 
Brimhall estimates the chances of a man of the generality 
getting into Who's Who as one in 823.5. For men of the aver- 
age age of the fathers of our gifted children the chances are 
probably less than one in 2,000. The actual number is three 
out of 578, which is about 10 times the number chance would 
give. In the case of women of the generality, Brimhall finds 
that the chances are only one in 7,647 of inclusion in Who's 
Who. Of 578 mothers of our gifted there are two of this dis- 
tinction, which, considering their average age of 37 years, is 
many, many times the number chance would give. Brimhall 
found that the father of a distinguished man of science is 98 
times as likely to be listed in Who's Who as is a man of the 
generality, and the brother of a distinguished man of science 
about 72 times as likely. 

One of the fathers in Who's Who is also one of CattelFs 
1,000 most distinguished men of science, one parent is a 
college president, and one grandfather is a widely known 
man of science. The names of distinguished parents and 
grandparents are withheld to avoid identification of any of 
our subjects. 

Other relatives in Who's Who have been reported to the 
number of 35, which is probably below the actual. Four of 
these were first cousins of gifted children, two were aunts, 
and one an uncle. The entire list is given in Table 18, 

TABLE 18 

RELATIVES IN WHO'S WHO 

Name of Relative Relationship to Gifted Child Child 

1. Abbott, James F. Second cousin H. G. 

Prof, of Zoology, J O. 

Washington Univ., 

1904-17. Appointed 

commercial attach^ 

to American Embassy 

in Tokio. Member of 

staff of American 

Delegation to confer- 

ence on Limitation of 

Armaments, 1921. 

2. Bannard, Otto T. Uncle L- B. 

Banker. 

3. Bradford, Edward G. Great-uncle W. E. B. 

Judge* 



TRAITS OF GIFTED CHILDREN 



TABLE 18 Continued 



Name of Relative 

4. Clark, Champ 

Speaker of House. 

5. Day, Holrnan F. 

Author Up in Maine, 
etc. 

6. Day, James R. 

Chancellor, Syracuse 
University. Author, 
The Raid on Prop- 
erty, etc. 

7. Eaton, Amasa M. 

Lawyer. 

8. Einhorn, Max. 

Professor of Medicine, 
P. G. Hospital, New 
York. Inventor of in- 
struments, author of 
medical books. 

9. Elkus, Abram. 

Lawyer. 



Relationship to Gifted Child 
Third cousin, once removed 

Not stated 



Great-uncle 



Great-uncle 



First cousin 



I First cousin 

Also related to 
Second cousin 



First cousin 



Second cousin, once removed 



Child 

J.D, 
F.D. 

E.H. 



E.H. 



Rufus K, 
Roger K. 

V.R. 



R.E. 

M.L. 
K. L, 

M.M, 



CM.F. 

JI.F. 

E.V, 



10. Fick, Arthur D. 

Author, The Happy 
Princess, etc. 

11. Gilbreth, Mrs. Lillian First cousin, once removed E. B. 

Psychologist. 

12. Goldberg, Rube 

Cartoonist. 

13. Harvey, G. B. M. 

Editor, New York 
World, 1891, Ambas- 
sador to England, etc. 

14. Hillis, Newell Dwight Distant cousin 

Writer. 

15. Johnson, Hiram Great-uncle 

U. S. Senator. 

16. Keller, Lue Alice Aunt 

Musician, lecturer, 
composer. 

17. Lindley, Alfred Not stated 

Banker. Directed fi- 
nances for 10 states 
in presidential cam- 
paign, 1920. 



M. E, K. 



G.K, 



INTELLECTUALLY SUPERIOR RELATIVES $5 

TABLE 18 Continued 

Name of Relative Relationship to Gifted Child Child 

18. Lindiey, Judge Curtis Not stated G. K. 

Lawyer. 

19. Lindiey, Harvey Not stated O.K. 

Capitalist. 

20. Lindiey, Walter Not stated G. K. 

Physician and author. 

21. McCormick, S. B. Second cousin, once removed J. McC. 

Chancellor, Univ. of 
Pittsburgh. 

22. Martin, Clarence First cousin, once removed L. G. 

23. Michelson, Dr. Alhert Cousin R. E. 

Professor of Physics, 
Univ. of Chicago. 

24. Michelson, Miriam Second cousin R. E. 

Dramatic critic and 
short story writer. 

25. Peck, John Hudson Great-uncle T. W. 

President, Renssalaer 
Inst. 

26* Scarborough, Dorothy Second cousin, once removed H. K. 
Writer. Instructor in 
English, Columbia 
University. 

27. Scarborough, George Second cousin, once removed H. K. 

Playwright. 

28. Scarborough, L. R. Great-uncle H. K. 

Writer. President, 
Southwestern Theo- 
logical Seminary. 

29. Sproule, Wm. Second cousin R. 0. 

President, S. P. R. R. 
Company. 

30. Upshaw, W. D. Second cousin H. M. 

Congressman from 
Georgia. Writer. 

31. White, Frank Second cousin H.M. 

Ex-governor of North 
Dakota. United States 
Treasurer. 

32. White, Joseph Second cousin, twice removed E. V. 

Sanitarian. 

33. Yeziersha, Anzie Aunt V. R. 

Author Hungry 
Hearts, etc. 



96 TRAITS OF GIFTED CHILDREN 

TABLE 18 Concluded 

Name of Relative Relationship to Gifted Child Child 

34. Young, James Addison First cousin, twice removed K. K. 

Supreme Court Justice. 

35. De Young, M. H. Great-uncle R. E. 

Founder of Golden 
Gate Museum. 

OTHER RELATIVES OF DISTINCTION 

There were 58 other relatives reported (in the Home 
Blank) of sufficient distinction to be named in standard 
cyclopedias of biography. These are listed below. Many 
relatives of perhaps equal distinction were excluded because 
no reference to them could be found in the biographical 
cyclopedias in the Stanford University Library. The list 
would probably have been considerably longer if foreign 
biographical cyclopedias had been more extensively con- 
sulted. Among the 58 are six signers of the Declaration of 
Independence, two presidents of the United States, two vice- 
presidents, four governors of states or colonies, four gen- 
erals, six writers, two inventors, four statesmen, three artists, 
and two judges of supreme courts. 

In Table 19 "Appleton" refers to Appleton's Cyclo- 
pedia of American Biography, and "National" to the Na- 
tional Cyclopedia of American Biography. In each case the 
relationship is given as reported in the Home Blank, and 
allowance must be made for ocasional inaccuracy, especially 
in the designation of cousin relationships and in the number 
of "great" prefixes. The latter suffix to a child's number indi- 
cates that the child has one or more sibs (indicated by an- 
other letter suffix) in the gifted group. 

TABLE 19 
OTHER RELATIVES OF DISTINCTION 

Child or 
Children to 
Name of Relative Relationship to Gifted Child whom 

1. Abbott, Lyman Distant cousin S. A. 

Clergyman and writer. J. A* 

Appleton, D, A* 

D.A. 

2. Alden, John Direct line A. M* 

Colonial fame. 
National. 



INTELLECTUALLY SUPERIOR RELATIVES 



97 



TABLE 19 Continued 



Relationship to Gifted Child 

Gt.-gt.-gt.-grandfather 
Direct line 



Name of Relative 

3. Alexander, Abraham 

Statesman. Appleton. 

4. Allen, Ethan 

Colonial fame. 
National. 

5. Anderson, Major Robt. Third cousin 

Commander at Ft. 
Sumter. Appleton. 

6. Babbitt, Isaac 

Inventor of Babbitt 
metal. Appleton. 



Gt.-gt.-uncle 



Not stated 



8. 



Gt.-gt.-grandfather 



Barnum, P, T. 

Founder of Barnum's 

circus and writer. 

Appleton. 
Bayrhoffer, Karl 

Philosopher, politi- 

cian and writer. 

Cassell's Biog. Diet. 
9. Bouguereau, Adolphe Gt.-gt.-uncle 

Painter; winner of 

premier grand prix 

de Rome. Siret's 

Diet, des Peintres. 

10. Bradford, Governor 

William 

Second governor of 
Plymouth Colony, 
Appleton. 

11. Bradstreet, Anne 

Colonial fame. 
National, 

12. Buchanan, James 

President of U. S. 
Appleton. 



Direct line 



Direct line 



Not stated 



Child or 

Children to 

whom related 

J.M. 
M.M. 
L.K. 
R.K. 

L.K. 
R.K. 

G.B. 
M.B. 
B.B. 
M.B. 
W.B. 



H.T. 
O.T. 



F.D. 
J.D. 



W. E. B. 



F.C. 



13. 



14. 



Buell, Gen. D. C. 

Civil War fame. 

Appleton. 
Carroll of Carrollton, Direct line 

Charles 

Signer of Declaration 

of Independence. 

Appleton. 



Distant cousin 

Not stated 

Second cousin, once removed 



J.F. 

R.H. 
D.C. 
G.M. 



S.A. 
J.A. 
D.A. 
D. A. 



98 



TRAITS OF GIFTED CHILDREN 



TABLE 19 Continued 



Name of Kelative 

15. Cartwright, Rt. Hon. 

Sir Richard 
Canadian Statesman. 
Canadian Men and 
Women of the Time. 

16. Cornell, Ezra 

Founder of Cornell 
University. Appleton. 

17. Cushman, Robert 

Secured grant for 
Plymouth Colony. 
Appleton. 

18. Dodge, Mary Mapes 

"Writer of children's 
stories and poems. 
Lippincott's Bio, Diet. 

19. Fairbanks, Charles 

Vice-president of U. S. 
National. 

20. Gerry, Elbridge 

Signer of Declaration 
and Vice-president of 
U. S. Appleton. 

21. Harrison, William 

Henry 

President of U. S. 
National. 

22. Healy, James 

Roman Catholic 
Archbishop; humorist, 
Diet., Universal Biog. 

23. Heyward, Thomas 

Signer of Declaration, 
Appleton. 

24. Hillier, Sir Walter 

K.C.M.G. and C.B. 
Author and inventor. 
Appleton. 

25. Hooker, Thomas 

First Governor, 
Connecticut Colony. 
Appleton* 

26. Hopkins, Stephen 

Signer of Declaration. 
Appleton. 



Relationship to Gifted Child 

Grandfather 



Child or 

Children to 

whom related 

R.C. 
P.C. 



Third cousin, once removed 
Direct line 

Maternal aunt 

Direct line 
Direct line 

Gt.-gt.-uncle 
Second cousin 

Not stated 
Grandfather 

Direct line 
Gt.-gt.-gt.-uncle 



K.B. 
R.N. 



M. G. 

S.G, 



B.D. 



S.A, 
J.A. 
D.A. 
D.A. 
R.B. 



M.K. 

W. E. B. 
R.H. 

F.F. 
A, McM. 



INTELLECTUALLY SUPERIOR RELATIVES 



99 



TABLE 19 Continued 



Name of Relative 

27. Johnston, Gen. 

Joseph E. 

General in Civil War. 

National. 

28. Kingsley, Charles 

Writer. National. 

29. Lee, Henry 

Light-horse Harry of 
Revolution. Appleton. 

30. Lee, Richard Henry Direct line 

Signer of Declaration. 
(Note: Henry Lee was 
a nephew of Richard 
Henry Lee. Robert E. 
Lee comes from the 
same family.) 
Appleton, 

31. Lewis, Francis Direct line 

Signer of Declaration. 
Appleton. 



Relationship to Gifted Child 
Second cousin 



Not stated 



Not stated 



32. Lewis, Morgan 

Captain in Washing- 
ton's army, and gov- 
ernor of New York, 
National. 

33. McCreery, John 

Poet. Appleton. 

34. Mapes, James 

Chemist, lecturer, 
editor and publisher. 
Appleton. 

35. Maxwell, Augustus 

Lawyer, railroad 
president, etc. Con- 
gressional Directory. 

36. Medley, Samuel 

Poet and divine of 
very distinguished 
ancestry. National. 

37. Morgan, Daniel 

Hero of Revolution. 
Appleton. 



Gt.-gt.-gt.-uncle 



Gt.-gt.-grandfather 



Gt.-grandfather 



Gt.-gt.-uncle 



Gt.-gt.-grandfather 



Direct line 



Child or 

Children to 

whom related 

C. M. 



R.B. 
D.D. 

T.C. 



J.I. 



J.I. 



E. M. 

S. G. 
M.G. 



C.M. 



E.L. 



C.C. 



100 



TRAITS OF GIFTED CHILDREN 
TABLE 19 Continued 



Name of Relative 

38. Mott, Lucretia 

Quakeress preacher 
and advocate of 
women's rights. 
Notable Women of 
History. 

39. Nevin, Robert 

Rector of St. Paul's 
Am. Church, Rome. 
Appleton. 

40. Opie, John 

Portrait painter. 
Member Royal 
Academy. National. 

41. Orr, James L. 

Statesman. Appleton, 

42. Osbourne, Frederick 

Canadian lecturer. 
Canadian Men and 
Women of the Time. 

43. Paroni, Giovanni 

Theologian and 
writer. Encyc. Brit. 

44. Patterson, Robert E. 

Governor of Pennsyl- 
vania. Appleton, 

45. Perkins, Samuel Elliott Not stated 

Judge of Supreme 
Court, Indiana, and 
Professor of Law. 
Appleton. 



Relationship to Gifted Child 

Not stated 



Gt.-grandfather 
Not stated 

Not stated 
Not stated 

Not stated 
Not stated 



46. Pierson, Abraham 

First head of Yale. 
National. 

47. Prentice, George D. 
' Poet and editor. 

Founded Louisville 
Journal. Appleton. 

48. Rolfe, John and 

Pocahontas 
Colonial fame. 
Appleton. 



Not stated 
Gt,-gt~gt.-grandfather 

Not stated 
Direct line 



Child or 

Children to 

whom related 

H. H, 



K.M. 
D.C. 

D,M. 
E. B. 

E.A. 
C. McC 



J.A. 
M.A, 



R.K. 
R.K. 

P.S. 



|F,D". 



LC. 



INTELLECTUALLY SUPERIOR RELATIVES 
TABLE 19 Concluded 



101 



Name of Relative 

49. Scott, WinfieJd 

General in Mexican 
War. Appleton, 

50. Sewall, Samuel 

Appleton. 

51. Snow, Lorenzo 

President of Mormon 
Church (1898). 
Ency. Brit. 

52. Stanton, Edwin M. 

Statesman. Appleton. 

53. Trumbull, Jonathan 

Statesman. Appleton 

54. Ward, Genevieve Distant cousin 



Relationship to Gifted Child 

Not stated 



Direct line 
Gt.-grandfather 



Gt.-gt.-uncle 
Direct line 



55. West, Benjamin 

Artist. Member of 
Royal Academy. 
Champlin's Cycl. of 
Art and Artists. 

56. Wilson, Thomas 

Judge, Supreme 
Court; Senator from 
Minnesota, Appleton. 

57. Winthrop, John 

Colonial fame. 
Appleton. 



Direct line 



Second cousin 



Direct line 



Child or 
Children to 
\vhom related 

V. McL 



R.K. 
R.K. 
M. McA. 



E.W. 

J.A. 
M.A. 
E. McM. 
E. McM. 
P.P. 



C. C. 
E.G. 



S.A. 
J.A. 
D.A. 
D.A. 

POSITIONS OF HONOR, TRUST, OR RESPONSIBILITY HELD BY 
PARENTS AND GRANDPARENTS 

The data summarized in Table 20 are from the Home 
Blanks which had been returned by 528 of the 578 families in 
the main gifted group at the time of tabulation. The list is 
known to be very incomplete, due to the frequent failure of 
parents to report honors of considerable importance. The 
positions of honor reported have been summarized under the 
headings political, religious or fraternal, professional or 
academic, business or financial, and miscellaneous. In many 
cases two or more positions of honor were named for the 
same parent or grandparent. Unfortunately, at the time the 
tabulations were made, duplicate positions of honor were 



102 TRAITS OF GIFTED CHILDREN 

not separately recorded; accordingly, the totals for each 
group refer to total number of honors, and not to number of 
parents or grandparents who have attained such honors. It is 
evident, however, that the proportion of gifted ancestors in 
this group of 528 families is extraordinarily high in compari- 
son with the generality. 

TABLE 20 

POSITIONS OF HONOR, TRUST, OR RESPONSIBILITY HELD BY 
PARENTS AND GRANDPARENTS 

Grand- Grand- 
Father Mother fattier mother Total 

A. Political: 

1. Major national government 
offices, as senator, represen- 
tative, etc. 11 9 20 

2. Major state offices as above 8 17 1 26 

3. Mayor of city of 25,000 or 

more 20204 

4. Mayor of city or town, less 

than 25,000 1 11 12 

5. Important civic offices other 

than mayor 10 1 26 37 

6. Minor political offices 32 18 93 2 145 

Total, Group A 64 19 158 3 244 

B. Religious or fraternal orders: 

1. Pres. or vice-pres. state or 
large city branch of orders, 

such as DAR, KC, etc. 14 2 13 10 39 

2. Important church offices, 
(bishop, etc., over large 

areas) 8 5 15 28 

3. Miscellaneous club offices 105 94 73 64 336 

4. Miscellaneous church offices 68 53 90 35 246 

5. Class offices in college 19 15 34 

Total, Group B 214 169 191 109 683 

C. Professional or academic: 

1. College president 02204 

2. College trustee 1 9 10 

3. Teacher in college 13 4 4 2 23 

4. School principal or 

superintendent 4 3 5 1 13 

5. Miscellaneous professional 

honors 55 16 19 90 

Total, Group C 73 25 39 3 140 



INTELLECTUALLY SUPERIOR RELATIVES 103 

TABLE 20 Concluded 

Grand- Grand- 
Father Mother father mother Total 

D. Business or financial: 

1. SupU or manager of large 

factory or corporation 46 28 74 

2. Bank president 4 14 18 

3. Bank director or trustee 10416 

Total, Group D 51 46 1 98 

ZL Miscellaneous: 

1. Commissioned officer, army 

or navy 17 27 44 

2. Public speaker of note 00011 

Total, Group E 17 27 1 45 

INCOMPLETE LIST OF DISTINGUISHED RELATIVES OF THREE 

GIFTED SIBS 

We have not thus far been able to undertake the comple- 
tion of any genealogies of our gifted families, although con- 
siderable material has been assembled for several. With 
some hesitation, owing to the incompleteness of our informa- 
tion, the available data are presented for one family. This 
family is doubtless above the average of our group, both as 
to number of distinguished relatives and as to completeness 
of family records; nevertheless, there are many other fam- 
ilies in the group, which, judging from such information as 
we have, would make almost or quite as good a showing. 

Most of the information regarding this family was ob- 
tained by Miss Goodenough from written or printed records, 
including genealogies, D.A.R, records, and Apple ton's Cyclo- 
pedia of American Biography. On the Home Blank only 
seven distinguished relatives were mentioned. Follow-up of 
the case brought to light 34 relatives sufficiently distin- 
guished to be named in Appleton's Cyclopedia of American 
Biography or in Who's Who, besides many others somewhat 
less distinguished. Of course, a majority of these relatives 
are too remote to be of much significance, individually, in 
the biological make-up of the propositi, but the significant 
fact is the frequency with which gifted relatives appear. The 
family shows a high order of selective mating throughout. In 
the following (incomplete) list of distinguished relatives 
those designated by a star are named in Appleton's Cycle- 



104 TRAITS OF GIFTED CHILDREN 

pedia or Who's Who. The names are grouped according to 
the generation to which they belong, counting back from the 
propositi. In order to preserve the anonymity of the imme- 
diate family, all names in the direct line are omitted in the 
first three generations. 

TABLE 21 

SUPERIOR RELATIVES OF A GIFTED FAMILY 
Zero generation. 

Propositi: IQ 183; IQ 140; IQ 131, sibs. 
First generation. 

1. Father. Graduate of Stanford and of Harvard Law School. 

Very successful attorney. 

2. Father's brother. Bond buyer. 

3. Mother. Amateur artist of some ability. 

4. Mother's brother. Inventor. 

5. Mother's brother. Mining engineer, 

Second generation. 

6. Paternal grandmother. Speaks "several" languages. 
*7. Maternal grandfather. Artist and collector. 

*8. Brother of 7, Lawyer. 

*9. Joseph Bartine. Remote cousin of paternal grandfather. Naval 
officer, commander of Philadelphia Navy Yard, Inspector 
of Lighthouses, and President of Naval Examining Board. 

Third generation. 

*10. Paternal great-grandfather. Author of several religious books 

and many poems. 

11. Paternal great-grandfather. Wall street financier, 
*12. Rufus King. Remote cousin of paternal great-grandfather, 

Graduate of Bowdoin. Writer on religious and historical 

subjects. 
*13. John Smith. Cousin of 12. Professor rhetoric and English 

literature at Bowdoin. 
*14. Isaac Hull, Remote cousin of direct ancestor* Naval oJBEIcer 

and commander of the Constitution. 

*15. John Carter Brown II. Second cousin of direct ancestor- 
Merchant, philanthropist, and bibliophile. Gave $160,000 

to erect a university library. 

Fourth generation, 

*16. Samuel Sewall IV. Cousin of direct ancestor. Clergyman and 
writer of local histories. 

*17. Nicholas Brown ,11. Cousin of direct ancestor. Member of 
Rhode Island legislature and one of leading patrons of 
Rhode Island College, the name of which was changed to 
Brown University in his honor. Founder of many worthy 
charities, including a hospital tor the insane at Providence, 

* In Appleton*s Cyclopedia or Who's Who. 



INTELLECTUALLY SUPERIOR RELATIVES 105 

TABLE 21 Continued 

*18. Obadiah Brown. Cousin of direct ancestor. Merchant, manu- 
facturer, and philanthropist. One of the firm of Almy, 
Brown and Slater, which first introduced cotton spinning 
into this country. Left endowment of $100,000 to the 
Friends Boarding School founded hy his father. 

*19. William Hull. Remote cousin of direct ancestor. Graduate of 
Yale. Major in Washington's army. Personally honored by 
Washington for his services at Stony Point. General in 
War of 1812. 

Fifth generation. 

*20. Samuel Sewall III. Brother of direct ancestor. Chief justice 
of Massachusetts supreme court. Member of Congress for 
two terms. 

*21. Thomas Sewall. Remote cousin of 16. Physician. Author of 
Pathology of Drunkenness, etc. 

*22. Jotham Sewall. Remote cousin of 16. Evangelist. 

*23. John Brown II. Direct ancestor. One of "Four Brothers 
Brown." Wealthy merchant and trader. His ships went to 
all parts of the world. Member of Congress, 1799. One of 
the chief patrons of Rhode Island College. 

*24. Nicholas Brown. Merchant and philanthropist. Patron of 
Rhode Island College. 

*25. Joseph Brown. Scientist and inventor, especially interested 
in electricity and astronomy. Held chair of natural philos- 
ophy in Rhode Island College, serving without compensa- 
tion. 

*26. Moses Brown. Founded the Friends Boarding School in 
Providence. Philanthropist and leader in Society of Friends 
of Rhode Island. 

Sixth generation. 

*27. Stephen Sewall I. Second cousin of direct ancestor. Hebrew 
scholar. Prolific writer and translator. 

*28. Samuel Sewall V. Inventor. First to drive piles as foundation 
for bridges. 

*29- David Sewall. Justice of superior court, and member of Massa- 
chusetts council. 

*30. Jonathan Sewall. Second cousin of direct ancestor. Lawyer, 
orator, writer. 

*31. Jonathan Mitchell. Second cousin of direct ancestor. Poet. 

32. Edmond Kimball. Direct ancestor. Collector of taxes under 

King's Commission in colonial Massachusetts. 

33. Asa Buel. Direct ancestor. Captain in Continental Army. 

Member of Connecticut State Legislature in 1802. His niece, 
Mrs. Sara Josepha Hale, was editor of Godey's Lady's Book. 
*34. James Brown II. Direct ancestor. Leading merchant in 
Providence and father of the "Four Brothers Brown." 



106 TRAITS OF GIFTED CHILDREN 

TABLE 21 Concluded 
Seventh generation. 

*35. Stephen Sewall L Cousin of direct ancestor. Judge of Su- 
preme Court of Massachusetts. 
*36. Joseph Sewall. Direct ancestor-. Son of Samuel Sewall. Elected 

president of Harvard in 1724 but declined. 

37. Samuel Kimball. Direct ancestor. First overseer Merrimack, 
afterward Bradford, Massachusetts. Held the rank of a cor- 
net of a troop of horse. 

*38. James Brown I. Direct ancestor. Pastor of Baptist Church in 
Providence and leader in the colony. 

Eighth generation. 

*39. Samuel Sewall I. Direct ancestor. Famous judge in Salem 
witchcraft trials. (The only one of the judges who later 
publicly admitted his error.) Entered Harvard at 15. 
40. Abraham Pierson III. Direct ancestor. Known as the 
"Worshipful.** Magistrate for fifty years in Clinton, Con- 
necticut, 

*41. John Brown I. Direct ancfestor. Member of town council of 
Providence. Leader in colonial affairs. 

Ninth generation, 

*42. John Hull. Direct ancestor. Goldsmith. Mintmaster and 
treasurer of Massachusetts colony. 

43. Richard Hull. Man of importance in Massachusetts colony. 

44. Richard Kimball. Direct ancestor. Puritan, Came over in 

the "Elizabeth" 1634. 

45. Samuel Buell. Direct ancestor. One of the original settlers in 

Killingworth, Connecticut. Owned estate of 1000 acres. 

Was styled "gentleman.** Commissioner of the Peace for 

several years, 
*46. Abraham Pierson II. Direct ancestor. First president and 

one of ten founders of Yale. 
*47. Chad Brown. Direct ancestor. One of the founders of Rhode 

Island Colony. Succeeded Roger Williams as leader. Was 

called "The Peacemaker," 

Tenth generation. 

48. William Buell, Direct ancestor. Emigrated from England In 
1630. One of the original settlers in Windsor, Connecticut, 
which he assisted in laying out 

*49. Abraham Pierson. Direct ancestor. Graduated from Cam- 
bridge. Emigrated to America in 1839. Pastor of Boston 
Church. Author of several religious books, Help for In- 
dian$> etc. 

Other distinguished relatives of this same propositi include 
Roger Williams, Ezekiel Howe, and Benjamin Franklin, but we do 
not have the data regarding exact relationships. 



INTELLECTUALLY SUPERIOR RELATIVES 107 

THE A FAMILY 

The A family is the result of a Japanese-American 

marriage. There are five children, all but one of whom have 
qualified for our gifted group. That one is still too young to 
test. 

Girl age 12 eighth grade IQ 137 
Boy age 11 eighth grade IQ 150 
Girl age 9 fifth grade IQ 140 
Boy age 7 second grade IQ 147 
Girl age 1 has not been tested. 

This is indeed a remarkable family, and the fact that it is 
the result of a mixed marriage makes it doubly interesting. 
This was the first, or one of the first, Japanese-American 
marriages in America and it aroused much unfavorable com- 
ment at the time. The parents were married in Portland, 
Oregon, because of the difficulty of finding any one in Cali- 
fornia who would consent to perform the ceremony. Follow- 
ing are a few facts regarding the heredity of the children. 

Paternal. The father of the propositi was born in Japan. 
He had four years of high school education, later studied 
silk culture, and before coming to America lectured on this 
and other subjects. In this country he has followed several 
occupations, but is now a florist and nurseryman. His father 
was born in Japan, of Japanese parents, and was educated 
by a private tutor. He was a fencing master and producer of 
silkworm eggs. He was very fond of reading, was a good 
chess player, and was twice elected Ko-cho, or head of the 
town. Little is known of the paternal grandmother of the 
children except that she was born in Japan, of Japanese 
parents, and that she was educated by a private tutor. 

Maternal. The mother was born in California and comes 
of exceptionally superior colonial ancestry. She attended 
college one year. She has held important offices in local 
churches and clubs and has published several poems and 
essays in newspapers and magazines. Her mother was born 
in California, had a high school education, is prominent in 
church and club circles, and lectures on home economics. 
She is descended from Charles Carroll of Carrollton, a 
"signer," and is related to Lyman Abbott. The maternal 
grandfather of the propositi was born in Massachusetts. He 



108 TRAITS OF GIFTED CHILDREN 

is a prominent man and has held many church offices of 
more than local importance. After the San Francisco earth- 
quake and fire he was an outstanding figure in the work of 
rehabilitation. In the direct line of his ancestry are six colo- 
nial governors (including John Winthrop and William Brad- 
ford), and Elbridge Gerry, a -'signer" and the fifth vice- 
president of the United States. There are also in the ancestry 
of the propositi several writers, artists, judges, clergymen, 
and government officials. 

We have another mixed Japanese-American family rep- 
resented in our group by a child of IQ 153, As in the A 

family, it is the father who is of Japanese descent. Each of 
the parents comes of rather superior stock. Still another 
child of Japanese-American parents is included in our spe- 
cial group showing artistic ability. In this case the mother is 
Japanese. In our outside Binet group, there is a child of 149 
IQ of mixed Chinese- American parentage (father Chinese) . 
All of these mixed marriages would have to be regarded, 
from a biological point of view, as highly successful. They 
illustrate the fact that in determining the quality of the off- 
spring the racial factor is much less important than the in- 
fluence of near ancestors. 

THE R FAMILY 

This is a Roumanian-Jewish family of eight children, 
three of whom belong to our gifted group : twin girls age 
eight fourth grade IQ's 142 and 131, and a girl age ten 
seventh grade IQ 139. A brother, IQ of 123 at age of four- 
teen, will complete high school at fifteen. He has been an 
honor student throughout his high school course and 
has exceptional musical talent. He has played the clarinet 
several times for the Los Angeles Examiner Radio Broad- 
casting Company and has composed orchestral pieces. His 
ten year old sister is also musical and sometimes accom- 
panies him in broadcasting. The school records of the other 
children in this family are all very superior. 

The parents came to America at about the age of twenty, 
practically penniless. The father worked at odd jobs and the 
mother in a factory. Neither parent had had more than a 
few months of schooling in Roumania. After coming to 
America the father attended night school, which the mother 



INTELLECTUALLY SUPERIOR RELATIVES 109 

was unable to do because of the children. When the oldest 
child began to attend school, the parents had him bring home 
his books in order that they might study with him of eve- 
nings. They have continued to educate themselves, now read 
and write English well, and are unusually well informed on 
most subjects of general interest. The mother, especially, 
has read a great deal along lines of child training and child 
welfare and shows remarkably good judgment in putting 
theory into practice. 

The family is still in very moderate financial circum- 
stances. The father worked in a butcher shop for several 
years, but failing health recently made it necessary for him 
to find lighter work. They accordingly rented a tiny shop 
next to their home (which is in a foreign neighborhood 
near the outskirts of the city) and have opened a small gen- 
eral store which the mother helps to look after. Their home 
contains only the barest necessities, as far as furnishings are 
concerned, but there is a piano and also a small library of 
well-chosen books, including the Book of Knowledge. Both 
parents appear to have decidedly superior intellectual abil- 
ity. Certainly their opportunities have been no better than 
those of thousands of foreigners with stupid children whose 
inferior intellectual ability is so commonly ascribed to lack 
of educational advantages. 

THE B FAMILY 

In the B family are four girls, all in the gifted group. 

One is a fifteen year old girl with a Terman Group Test score 
of 191. She is an "A" student in all her work and plays the 
cello and piano. 

The second is thirteen and has a Binet IQ of 137. She 
leads her class in the first year of high school and plays the 
cornet in the high school orchestra. 

The third is eleven years old and in the eighth grade. Her 
Binet IQ is 152. She plays the violin in the Los Angeles Chil- 
dren's Orchestra. Music teachers and musicians who have 
heard her play regard her as highly gifted musically. 

The fourth is nine years old and is in the sixth grade. Her 
IQ is 172. She learned to read at the age of four without in- 
struction. She plays the flute and was admitted to the Los 
Angeles Children's Orchestra after six months of instruction. 



110 TRAITS OF GIFTED CHILDREN 

The father of these children was manager of a sugar of 
milk factory and invented machinery for use in the factory, 
but he developed manic-depressive insanity at the age of 53 
and has since been confined in a sanitarium. The mother 
owns and manages a real estate office and supports the fam- 
ily. She had three years in college. The paternal grandfather 
was a teacher, of Scotch and French descent. The paternal 
grandmother, also of Scotch and French descent, was a 
woman of very superior intelligence, but was mentally un- 
balanced for a period of two years at the time of the meno- 
pause. The maternal grandfather was entirely self-educated 
but was a great reader and especially fond of Emerson and 
Carlyle. The maternal grandmother seems to have been ap- 
preciably above the average in intelligence. 

It is interesting that the four children of this family 
should combine exceptional musical ability with such 
marked intellectual superiority; also that the ancestry dur- 
ing the two preceding generations gave no evidence of very 
exceptional musical talent. However, the mother plays the 
piano and is well informed along musical lines* 

SUMMARY 

1. The 578 families of the main group have yielded 676 
subjects, although nearly 500 of their sibs have not been 
tested. Seventy-three families have yielded two subjects, and 
nine three or more. The number of families with two sub- 
jects is more than 1,200 times the number chance would give. 

2. Nearly a quarter of the members of the Hall of Fame 
are known to be related to one or more of our subjects. The 
number in our main group known to be so related is 15, or 
2.3 per cent of the entire number* 

3. Although a majority of the parents are relatively 
young, five are listed in Who's Who, three fathers and two 
mothers. This is many times the number chance would give. 
Four grandfathers and 35 other relatives are known to 
appear in Who's Who. 

4. Among 58 other eminent relatives are six signers of the 
Declaration of Independence, two presidents and two vice- 
presidents of the United States, four generals* six writers, 
two inventors, four statesmen, three artists, and two supreme 
court judges. 



INTELLECTUALLY SUPERIOR RELATIVES 111 

5. Parents and grandparents have held posts of responsi- 
bility in very great number, including 20 cases of major 
national office, 26 of major state office, 67 of major religious 
or fraternal office, 4 college presidencies, 23 professorships 
in colleges, 74 positions as superintendent or manager of a 
large factory or corporation, and 18 bank presidencies. 

6. There is one family in the group which has 34 known 
relatives sufficiently distinguished to be named in Appleton's 
Cyclopedia of American Biography or in Who's Who, be- 
sides many others somewhat less distinguished. The indica- 
tions are that many other families in the group have gene- 
alogies hardly less interesting. One of these is a family of 
five children (four of them in the gifted group), who are the 
offspring of a Japanese-American marriage. 

The data set forth in this chapter are very incomplete, but 
fragmentary as they are, they give considerable support to 
Galton's theory as to the hereditary nature of genius. Unfor- 
tunately, it has thus far not been possible to carry out any 
studies of a kind which would give exact data on family 
resemblances in the group or reveal the laws by which su- 
perior mental ability is transmitted. 



CHAPTER VI 

VITAL STATISTICS 

SIZE OF FAMILIES 

There are only 91 completed families in our group, if 
we count as completed a family in which the mother is 45 
years old or older. These 91 mothers reported to the medical 
examiners a total of 353 pregnancies, two of which were of 
twins. The number of children which might have resulted 
was therefore 355. Forty-six miscarriages and 45 deaths 
were reported. The 309 births exclusive of miscarriages 
include 165 boys and 144 girls, giving a sex ratio of 114.58 
to 100. 

In the same 91 completed families information was avail- 
able regarding the size of the family from which 85 of the 
fathers and 88 of the mothers had come. The correlation 
between number of children in the present generation and 
the number in each of the parents* families was as follows : 

Immediate family us. father's family, r = -K36 .06 
Immediate family vs. mother's family, r = -K18 .07 

The 91 families were distributed as follows with respect 
to the number of children in the present generation : 

Number of children 1 2 3 4 5 6 7 8 91011 
Number of families 19 22 16 11 7 7 3 2 2 .. 2 

These data yield an average number of children of 3.40 
to the family. At first thought such an average might be 
taken to indicate that the California gifted families are 
probably in general maintaining their numbers. Further 
consideration, however, shows that the average of 3.40 is 
subject to three corrections. 

First correction. There is a selection of families "caught" 
in our survey, due to the fact that a large family has a better 
chance of coming into our group than a small family. To 
take a simple hypothetical case, if half the marriages of a 

113 



114 TRAITS OF GIFTED CHILDREN 

population produced two children, and if the other half of 
the marriages produced four children, the average number 
of children per marriage, for the total population, is 3. How- 
ever, since the four-child marriages taken together produce 
twice as many children as the two-child marriages, it comes 
about that any random sampling of the children in such a 
population would give us twice as many four-child families 
as it would give of two-child families. For example, if 10 
two-child families were caught (20 children), the expected 
number of four-child families caught would be 20 (80 chil- 
dren). The sum of the children (100) divided by the number 
of marriages (30) gives the average number of children for 
the sampling as 3.33, which is to be compared with the 
average of 3.00 for the total population. Accordingly, any 
group of families located in the manner in which our gifted 
families were located will yield an average number of 
children which is spuriously high. As Dr. Lenz 1 has pointed 
out, "since the probability for each family to be included in 
the random sample is proportional to the number of its 
children, one must divide the number of families by the 
number of children in order to get the same composition 
of the material as in the totality of all fertile marriages." 
Table 22 gives the necessary computations. 

The fertility index of fertile marriages of this class of 
parents in California is not represented by 3.40 (the average 
number of children in the families caught), but by 2.18. 

Second correction. It must be borne in mind that the 
index 2.18 holds only for fruitful marriages. Since mar- 
riages which were childless could not be included in our 
sampling, a further correction is necessary. Assuming the 
proportion of childless marriages to be the same as that re- 
ported by Cattell for the generality of marriages in America, 
viz., 17 per cent, the 91 fruitful marriages indicate a total 
number of marriages of 110, and our fertility index of 2.18 
is accordingly reduced to 1.80. 91x2.18-110=1.80.) 

1 Dr. Fritz Lenz, Eugenical News, January, 1926, p. 2 ff. I am greatly 
indebted to Dr. Lenz for calling my attention to certain errors which found 
their way into the treatment of this section in the first edition of the book. 
The errors were due to failure to take account of childless marriages and 
of celibacy, and resulted in a material overestimate of the fertility of 
California gifted families. The present treatment follows that used by 
Dr. Lenz in the article referred to. 



VITAL STATISTICS 115 

Third correction. The index 1.80 is for those individuals 
who marry and have children, but since a large proportion 
of the individuals of this class remain celibate, a third 
correction must be computed. In all probability this pro- 
portion is as high as 20 per cent. If we base our estimate 
upon this figure, our index of 1.80 is reduced to 1.44. 

TABLE 22 
CORRECTING FERTILITY INDEX FOR METHOD OF SAMPLING 



1 


2 


3 


Number of 


Number of 




Children 


Families 


CoL 2 -r- Col. 1 


1 


19 


19.00 


2 


22 


11.00 


3 


16 


5.33 


4 


11 


2.75 


5 


7 


1.40 


6 


7 


1.17 


7 


3 


0.43 


8 


2 


0.25 


9 


2 


0.22 


10 





0.00 


11 


2 


0.18 


Sums 


91 


41.73 



91-*- 41.73 = 2.18 

(91x2.18138=:1.44.) Accordingly, if our data may be 
regarded as typical, it appears that a man or woman repre- 
senting the stratum with which we are here concerned is 
producing on the average .72 of a child, and that the gifted 
families of California are rapidly dying out. 

That the situation in the present generation is eugenically 
less favorable than in the preceding generation is shown by 
a comparison of the figures above with the corresponding 
figures for the fathers' and mothers' families. As we have 
already stated, data were available with respect to the size 
of the father's family in 85 of the 91 cases, and with respect 
to the size of the mother's family in 88 cases. The distri- 
butions are given in Table 23. 

TABLE 23 
SIZE OF FAMILY IN PRESENT AND PRECEDING GENERATIONS 

Number of children 1 2 3 4 5 6 7 8 9 10 11 12 14 19 
Fathers' families 7 8 12 12 10 9 10 7 3 3 1 2 1 

Mothers' families 1 8 12 12 11 14 12 5 5 1 5 2 



116 TRAITS OF GIFTED CHILDREN 

Carrying out the same computations for the parents' fam- 
ilies as we have already made for the families of the present 
generation, but allowing for 10 per cent of celibacy instead 
of 20 per cent, since it is known that celibacy in the educated 
classes was formerly less frequent than at present, we have 
the significant results shown in Table 24. 

TABLE 24 
FERTILITY INDICES OF PRESENT AND PRECEDING GENERATIONS 

Families Families Families 

of of of 

Children Fathers Mothers 

1. Number of children of selected mar- 

riages 3.40 5.33 5.66 

2. Estimated number for fruitful mar- 

riages in this stratum of popula- 
tion 2.18 3.44 4.37 

3. Estimated number for all marriages 

in this stratum of population 1.80 2.87 3.63 

4. Estimated number after correction 

for celibacy 1.44 2.59 3.26 

5. Estimated number per individual 0.72 1.28 1.63 

If we base our comparison on the estimated number of 
children for all marriages in this stratum of the general 
population (item 3 above), the reduction in fertility is from 
an average of 3.25 in the parental generation to 1.80 in the 
present generation. This is a decrease of 45 per cent. When 
we take account of the increasing celibacy (item 4, above), 
the reduction is from 2.92 in the parental generation to 1.44 
in the present generation. That is, the fertility index for the 
stratum of the California population with which we are 
here concerned has decreased by 50 per cent in a single 
generation. 

Because of the various corrections that need to be taken 
into account there are few data on fertility strictly compar- 
able to our own. Cattell 1 found the average number of 
children born to 440 American men of science, whose 
families were completed, to be 2.3. Barren marriages are 
included in this average, so that the figure 2.3 for this scien- 
tific group is to be compared with 1.80 (the figure resulting 
from the second correction) for the present generation of 

1 American Men of Science, New York, 1921, p. 793. 



VITAL STATISTICS 117 

our gifted families. Galton 1 found that a group of 100 
English men of science, excluding barren marriages, had on 
an average 4.7 children. Havelock Ellis 2 found an average 
of 5.45 children for 214 fruitful marriages of British men 
of distinction. The figures 4.7 and 5.45 are to be compared 
with the results of the first correction for the present gener- 
ation of gifted families, namely, 2.18. It would seem, there- 
fore, that the fertility of families of the type which produce 
gifted individuals is rapidly on the decline. 

SIZE OF FAMILY AND EDUCATION OF PARENTS 

An attempt was made to find out whether within this 
group of 91 families there was any correlation between the 
amount of schooling of parents and the number of offspring. 
The highest school grade reached was used as the measure 
of schooling, and a schooling index for mid-parent was cal- 
culated by averaging the highest grade reached by the two 
parents. For example, one parent reached Grade 12 and the 
other, Grade 8, the mid-parent index of schooling was 10. 
The correlation with number of offspring was found to be 
-.214+.07. 

SIZE OF FAMILY AND DEGREE OF SUPERIORITY OF CHILDREN 

Another question that arises is whether there is any cor- 
relation, within our highly selected group, between the de- 
gree of a child's superiority and the size of family from 
which he comes. The correlation was computed for the 91 
completed families and found to be .271 rt. 062. This cor- 
relation is high enough to be decidedly significant, from the 
point of view of eugenics, and it is desirable to know 
whether it can be accounted for on the hypothesis that the 
most gifted parents (those most likely to have children of 
highest IQ) are prevented from having large families by 
their greater educational ambitions. It is possible to answer 
this question by the method of partial correlations. The 
correlation between IQ of child and schooling of mid-parent 
was found to be +.16. The correlation between IQ of child 
and number of living births, with schooling of mid-parent 

1 English Men of Science, London, 1874; New York, 1875. 

2 A Study of British Genius, London, 1904. 



118 TRAITS OF GIFTED CHILDREN 

constant, is .246:.063. That is, the fact that the highest 
IQ's are found in the smallest families can not be accounted 
for by the supposed interference of schooling, as it is but 
slightly reduced when the effect of this factor is eliminated. 

MISCARRIAGES 

In the 91 completed families, 46 miscarriages were re- 
ported by the mothers to our examining physicians. The 
proportion of pregnancies which resulted in miscarriages, 
7.7 per cent, is very low. For 10,043 pregnancies, records of 
which were obtained in a hospital of Manchester, England, 
Dr. A. S. Parker 1 reports 1,659 abortions (miscarriages), or a 
proportion of 16.5 per cent. The proportion was very small 
in the younger mothers and rose to 20.7 per cent by the age 
of 28. Of mothers of gifted children, 57, or 62 per cent of the 
91, report no miscarriages, and in five cases the question is 
not answered. (In one of these the mother is dead. No rea- 
son for failure to reply in the other four cases is known.) 
The physicians asked in each case of miscarriage whether it 
was induced or spontaneous. It can be assumed that when- 
ever the information given by parents on this point is incor- 
rect, the error is pretty certainly in one direction, as there is 
little likelihood of a mother reporting a miscarriage as in- 
duced which was spontaneous. There are 23 who report 
spontaneous miscarriages only; one, induced miscarriages 
only; three, both induced and spontaneous; and three fail 
to state whether the miscarriage was induced or spontane- 
ous. The number of induced miscarriages admitted is nine. 
The following figures give the number and kind of mis- 
carriages for each mother: 

Number per mother 
1234 Total 

Mothers having spontaneous miscarriages 19 3 2 31 

Mothers having induced miscarriages 2 1 1* 9 

Kind not stated 111 6 

Total miscarriages 22 8 12 4 46 

* One mother reported four induced miscarriages and one that was spon- 
taneous, which gives one case of five miscarriages in the row of totals. 

If the chances of superior endowment were the same for 
these 46 aborted offspring as for their sibs, a considerable 

^Eugenics Review, Vol. 15, 1924, p. 584 ff. 



VITAL STATISTICS 119 

number of potentially gifted individuals have thus been lost 
to the world. In all probability some of these were lost 
through abortions which were induced. The total number of 
potentially gifted individuals thus sacrificed in a nation of a 
hundred million population must be very large, even when 
due allowance is made for the small proportion of gifted off- 
spring in the population at large. Probably a number of 
these mothers attempted miscarriages which they did not 
succeed in bringing about. One such failure is known to have 
given us one of our brightest and most promising boys. 

INFANT MORTALITY IN FAMILIES OF GIFTED 

The infant mortality in families of gifted children has 
been compared with that in the general population with re- 
spect to per cent of deaths under one year and under five 
years. This comparison has been made separately for the 91 
completed families for whom we have Home Blanks. The 
results given below show that infant mortality in our gifted 
families is about two-thirds of the expected in the case of 
boys, and only one-third of the expected in the case of girls. 
The good showing might indicate either better care, better 
heredity, or both. 

Percentage of Mortality 
Under 1 year Under 5 years 

Males Females Males Females 

General Population 12.32% 10.23% 17.03% 14.65% 

91 Completed Families of Gifted 8.8 % 3.4 % 11.9 % 5.5 % 

AGE OF PARENTS AT BIRTH OF GIFTED CHILD 

This question was asked in the Home Information Blank 
and was answered for 583 children, or 96.5 per cent of those 
for whom the blank was filled out. Accordingly, the data as 
shown in Table 25 may be considered representative. In 
this case there would seem to be no important selective fac- 
tor which would give an abnormally large number of chil- 
dren whose parents were especially old or especially young 
at the birth of their gifted offspring. Perhaps one selective 
factor does enter to a slight degree. Such would be the case, 
for example, in a population containing an abnormally large 
or an abnormally small proportion of young married 
couples. This proportion is likely to be rather large in fron- 



120 TRAITS OF GIFTED CHILDREN 

tier communities, but in cities such as we are here concerned 
with this factor would probably be small, even though they 
have had a fairly heavy immigration in recent years from 
other states. 

Cattell found the average age of fathers of 865 leading 
American men of science to be 35 years at the birth of their 
sons, and that of the mothers 29 years and 8 months. These 
figures agree very closely with those in Table 25. The 

TABLE 25 
AGE OF PARENTS AT BIRTH OF GIFTED CHILDREN 



Boys' Parents 


Girls' 


Parents 


Total Both Sexes 


Age 


No. 


Cases 


No. 


Cases 


No. Cases 




Fathers 


Mothers 


Fathers 


Mothers 


Fathers 


Mothers 


15-19 


1 


10 





5 


1 


15 


20-24 


27 


70 


29 


61 


56 


131 


25-29 


76 


119 


65 


90 


141 


209 


30-34 


105 


80 


71 


63 


176 


143 


35-39 


59 


31 


52 


30 


111 


61 


40-44 


33 


9 


24 


15 


57 


24 


45-49 


6 




14 




20 




50-54 


3 




7 




10 




55-59 


5 









5 




60-64 


3 




1 




4 




65-69 


1 




1 




2 




Total 


319 


319 


264 


264 


583 


583 


Mean 


33.63 yrs. 


28.74 yrs. 


33.64 yrs. 


29.34 yrs. 


33.63 yrs. 


29.01 yrs. 


Median 


32.64 yrs. 


28,35 yrs. 


32.68 yrs. 


28.65 yrs. 


32.66 yrs. 


28.48 yrs. 


S. D. 


7.68 yrs. 


5.56 yrs. 


7.78 yrs. 


5.87 yrs. 


7.70 yrs. 


5.64 yrs. 



average age of fathers of 299 British men of genius was 
found by Ellis to be 37.1. The corresponding figure for fath- 
ers of 100 English men of science studied by Galton was 36; 
for the mothers of this same group, 30. Ellis thinks there is a 
positive correlation between degree of eminence and age of 
father, but his data do not at all warrant this conclusion. The 
corresponding figure for the entire population would doubt- 
less be much lower than for any distinguished group, but the 
difference is probably accounted for entirely by age of mar- 
riage, voluntary limitation of births, and other factors hav- 
ing no biological basis. 

Table 25 shows that the average age of fathers is 4.62 
years in excess of the average age of mothers. The difference 



VITAL STATISTICS 121 

for parents of gifted boys is 4.89 years, and for parents of 
gifted girls, 4.30 years. The difference between medians is 
4.29 for parents of boys, 4.03 for parents of girls, and 4.18 for 
both. The difference between the excess of father's age over 
mother's for the gifted boy group as compared with the 
gifted girl group is well within the range of chance. The di- 
rection and extent of disparity in age of parents is shown in 
Table 26. 

TABLE 26 

DISPARITY IN AGES OF PARENTS 
(Minus sign indicates the father was younger than the mother) 





Boys' Parents 


Girls' Parents 


Total Both Sexes 


Years 


No. Cases 


No. Cases 


No. Cases 


10 and 9 


1 





1 


Sand 7 





2 


2 


6 and 5 


6 


3 


9 


4 and 3 


7 


7 


14 


2 and 1 


19 


25 


44 





31 


24 


55 


land 2 


56 


48 


104 


3 and 4 


63 


51 


114 


Sand 6 


47 


31 


78 


7 and 8 


29 


22 


51 


9 and 10 


24 


18 


42 


11 and 12 


11 


16 


27 


13 and 14 


7 


7 


14 


15 and 16 


5 


3 


8 


17 and 18 


3 





3 


19 and 20 


3 


1 


4 


21 and 22 


2 


4 


6 


23 and 24 


1 


1 


2 


25 and 26 


2 


1 


3 


27 and 28 


2 





2 



Total 319 264 583 

Statistics on the age of marriage of these parents would 
be interesting, but unfortunately this item was omitted by 
oversight from the Home Information Blank. 

ORDER OF BIRTH 

Owing to the form of our data it is impossible to work out 
the birth order exclusive of miscarriages, since the order of 
miscarriage is not stated and the order of birth of the pro- 
positi as stated includes miscarriages and deceased siblings. 



122 TRAITS OF GIFTED CHILDREN 

Another and more serious source of error in the accom- 
panying tables is the fact that, since our main survey was 
confined to the eight school grades, many children in the 
largest families of the general population were not caught in 
our group. Moreover, the tables include both completed and 
incompleted families, for to have based the tables upon the 
completed families alone would have favored the selection 
of late-order births. Accordingly, in many cases it is to be 
expected that later births will change the relative position of 
a given child in the table. Absolute position, of course, will 
not be changed, but a child who is now the oldest of three 
may become the oldest of four, five, etc. 

The tables should then be considered suggestive rather 
than conclusive, and this only as regards the small families; 
in the large families, as we have stated, the older children 
would often be beyond the age limits of our group. Never- 
theless the figures are interesting to compare with the corre- 
sponding figures from Cat tell. The line across the middle of 
Tables 27-29 is intended to call attention to the fact that the 
figures for birth orders beyond four are of questionable 
validity. 

The following figures give the per cent of birth of each 
order for families of 2, 3, or 4 children. CattelFs figures for 
American men of science are given in parentheses. 

ORDER OF BIRTH 

Children First Second Third Kourtli 

2 
3 

4 33.0% 26.8% 15.4% 24.7% 

------ (19.7%) 

There is a very striking agreement between Cattell's data 
and our own for families of two children. In each case nearly 
three-fifths are firstborn. The agreement is only moderately 
close for families of three and four children, which may be 
explained on the hypothesis that first born gifted in families 
of three or four children were more likely to have been 
missed because they had advanced beyond the eighth grade. 
It has been suggested that the better chance which the first 



First 


Second 


Third 


56.1% 

(57.4%) 


43.9% 
(42.6%) 




36.9% 
(44.0%) 


31.9% 
(31.2%) 


31.2% 
(24.8%) 


33.0% 
(36.1%) 


26.8% 
(22.4%) 


15.4% 

(21.8%) 



VITAL STATISTICS 



123 



TABLE 27 

ORDER OF BIRTH BOYS 
(Based on total number of pregnancies.) 

Order of Birth 



No. of 




















Children 


1 


2 


3 


4 


5 


6 


7 


8 


Total 


1 


49 


__ 





_ 


_ , 


__ 


_ 


, 


49 


2 


46 


39 


_ 





_ 


_ 


_ 





85 


3 


25 


26 


26 





_ 


_ 





_ 


77 


4 


18 


14 


5 


10 


- 


- 


- 


- 


47 


5 


7 


5* 


7 


4 


5 





_ 


_ 


28 


6 


2 


2 


3 


- 


1* 


4 





~ 


12 


7 


_ 


1 


1 


_ 








2 





4 


8 


_ 


_ 


_ 








_ 





1 


1 


9 


- 


- 


- 


- 


1 


1 


- 


- 


2 



Total 



147 87 + 1 42 14 7 



305 



* Includes one pair of twins, counted as one birth. (In 12 cases order was 
not stated.) 





TABLE 28 






ORDER OF BIRTH GIRLS 






(Based on total number of pregnancies.) 






Order of Birth 




No. of 






Children 


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 


Total 


1 


qq 


QQ 


2 


41 29 - .- -_ - - - 


70 


3 


27 19 18 _________ 


64 


4 


14 12 10 14 -______-._ 


50 


5 


43844 -___-._-__ 


23 


6 


1213-1 __-.--._.___ 


8 


7 


-11111 ____.__-.-_ 


5 




1 


1 


Q 


1 2 


3 


1 n 


j 


1 






o 


19 




o 


13 


__ _ __ ______ _ __ i ___ _ _ _ 


1 


14 


2* 


2 


1 ^ 




Q 


1fi 




o 






o 


18 


__ -_ 1-1- 


2 



Total 126 66 38 22 6 2 1 1 2 1 2 1 1 269 

* Includes one pair of twins, counted as one birth. (In five cases order was 
not stated.) 



124 



TRAITS OF GIFTED CHILDREN 



born apparently has of becoming a leading man of science 
may be due to greater educational opportunities enjoyed by 
thefirst born. This alleged educational advantage is, of course, 
hypothetical but, even if it were actual, one might very well 
doubt whether the effect would register in the Binet IQ to 
almost exactly the same degree as in scientific performance, 

TABLE 29 

ORDER OF BIRTH SEXES COMBINED 
(Based on total number of pregnancies.) 

Order of Birth 

Children 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 JSxotai 
! 88 _ _ ~ ~ J! 

2 87 68 - - ---------155 

3 52 45 44 - ---------141 

4 32 26 15 24 97 

5 11 8* 15 8 9 ---------51 

6 3 4 431* 5 20 

7 - 2 21112----------- 9 

8 - - __i_-i-_-.------- 2 

9 - - --111-2--------- 5 

10 - - -----I---------- 1 

n _ _ o 

12 - 

13 _ _ 1 1 

14 _ _--2*----- 2 

15 _ _ 

16 - - ___ 

17 _ _ _ 

18 - 1 - 1 - 2 

Total 273 153 80 36 13 7 3 2 2 1 2 1 1 574 

* Includes one pair of twins, counted as one birth. (In 17 cases order was 
not stated.) 

especially since the IQ's were taken in childhood, while the 
supposed influence of environment upon scientific perform- 
ance would presumably depend chiefly on the relative 
amount of schooling enjoyed by first born and later born 
after the years of childhood. Such facts seem to render the 
environment hypothesis very questionable. 

However, before accepting the above findings it would be 
necessary to rule out all factors tending to bring about an 
atypical selection of cases, and this is extremely difficult if 
not impossible. The misleading effect of the selective factor 
is illustrated by considering the data for the 91 completed 



VITAL STATISTICS 125 

families. It might at first be supposed that the data for these 
families would, except for the small number, be more valid 
than for the entire group. Actually they have no validity at 
all. The completed families in our group show an atypically 
small proportion of gifted first born, for the reason that our 
main search, being confined chiefly to grades 3 to 8, yielded 
few children above 13 years of age. The eldest child in a 
completed family would frequently, if not in a majority of 
cases, be older than 13, and would therefore be less likely to 
be caught in our survey. The enormous effect of this vitiat- 
ing factor is shown by the fact that in 24 families of this 
group having two children, the ratio of second-order to first- 
order births was 17 to 7. In 17 families of three children, 
there were 12 third-order births and only one of first order. 
The fact that a selective factor could give such results 
illustrates the necessity of caution in interpreting data of 
this kind. In this case completed families having a gifted 
first born were rarely caught. Possibly in the case of incom- 
plete families, those with gifted first born were more likely 
to be caught. Such would certainly tend to be the case with 
the families most incomplete, since the gifted second born 
of the youngest parents would not yet have reached the 
third grade. However, it will be noted that our data for 
families of two and three children agree very closely with 
those of Cattell, and it does not seem probable that CattelPs 
figures could be subject to the kind of error we are con- 
sidering. 

MORTALITY STATISTICS REGARDING PARENTS 

In the case of 502 of the 578 families of our main gifted 
group, including 591 of the total 643 children in the group, 
the physicians made note of whether the parents were living 
or dead. This includes all the families of the main group in 
which the children were given a medical examination. 
Among these 591 children, 2 (0,33 per cent) were orphans 
and 53 (8.9 per cent) were half orphans. With the number 
of half orphans at about 9 per cent, the number of full or- 
phans expected by chance would be about 0.81 per cent. 
(.09 X. 09=. 0081). This is not far from the proportion found. 

Strange -to say, the number of deceased fathers reported 
is more than three times that of deceased mothers; namely, 



126 TRAITS OF GIFTED CHILDREN 

42 fathers as compared with 13 mothers. These numbers are, 
respectively, 8.37 per cent and 2.6 per cent of the total 
number of parent pairs (502), a difference large enough to 
raise considerable question of accuracy. Two possible sources 

TABLE 30 
CAUSES OF DEATHS OF PARENTS 

Father Mother 

Accident 7 1 

Unknown 3 2 

Tuberculosis 3 2 

Influenza-pneumonia 4 1 

Heart disease 4 

Influenza 2 1 

Appendicitis 2 

Pernicious anaemia 2 

Nervous breakdown 2 

Diabetes 1 1 

Septicemia 1 1 

Pneumonia 1 1 

Hernia operation 1 

Peritonitis 1 

Cancer of liver 1 

Nephritis 1 

Meningitis or Encephalitis 1 

Sciatica 1 

Tuberculosis of kidney 1 

Disease of gall bladder 1 

Nephrectomy 1 

Childbirth 1 

Gangrene 1 

Suicide 1 

Old age 1 

Total 42 13 

of error suggest themselves. (1) The reports were made in 
nearly all cases by the mothers when they brought the child 
for medical examination. It is possible that some of these 
were stepmothers but did not divulge the fact. One might 
assume that the death of a father would always be reported 
by the mother; one could not safely assume that every 
stepmother would report that she was a second wife and 
that she had nothing to do with bringing into the world the 
child found to be so exceptionally superior! (2) A second 
source of error lies in the possibility that among the 75 or 80 
mothers who could have brought the child for a medical ex- 



VITAL STATISTICS 127 

animation and failed to do so, there was a disproportionate 
number of stepmothers. This source of error is probably 
less serious than the first. If neither of these errors has en- 
tered, and the excess of widowed mothers is a true one, the 
explanation might be that children of widowed mothers 
stood a better chance of being caught in our survey because 
of extra attention or instruction in the home. One might ex- 
pect widowed mothers to lavish more affection upon their 
offspring, but as a class they certainly do not have more time 
to devote to them. 1 

The causes of death in the case of these 42 fathers and 13 
mothers were reported (to our physicians) as shown above 
in Table 30. 

CHRONIC ILLNESSES OF PARENTS 

Chronic illnesses of parents were reported to our phy- 
sicians as shown in Table 31. The numbers are, of 
course, too small to be statistically significant, even if one 
could assume entire accuracy of report. One notes, however, 
that the mothers report 50 per cent more illnesses for them- 
selves than for their husbands! These illnesses do not in- 
clude tuberculosis and lues (syphilis), which were reported 
separately, 

TABLE 31 
CHRONIC ILLNESSES OF 502 PARENTS 

Father Mother 

Nervous troubles 7 12 

Hay Fever 6 7 

Asthma 3 5 

Goiter 8 

Deafness 5 

Heart disease 4 

Rheumatism 4 

Encephalitis 1 1 

Cancer 2 

Insanity 1 1 

Neurasthenia 1 1 

Neuritis 1 1 

Dyspepsia 1 1 

Lead poisoning 1 1 

Pleurisy 1 1 

Kidney trouble 1 1 

J A part of the discrepancy is due to the greater age of fathers. 



128 TRAITS OF GIFTED CHILDREN 

TABLE 31 Concluded 

Father Mother 

Urticaria 1 

Chronic indigestion 1 
Paralysis and Hodgkins 

disease 1 
Low blood pressure and 

ansemia 1 

Gall stones 1 

Otitis 1 

Mastoid 1 
Locomotor ataxia and 

alcoholic neuritis 1 

Arthritis (rheumatoid) 1 

Palsy 1 

Gastric ulcers 1 

Hyperthyroidism 1 

Hyperthyroidism (suspected) 1 

Diabetes mellitus 1 

Tumor 1 

Vaginal discharge 1 

Prolapsus 1 

Pelvic trouble 1 

Dementia prsecox 1 

Cardiac asthma 1 

Recurring abscesses 1 

High blood pressure 1 

Cataract 1 

Scoliosis 1 

Disease of gall bladder 1 

Illness (not stated) 1 

Total 40 65 

Minus duplications 4 

40 61 

Duplications: 1 mother, goiter and hyperthyroidism 

1 mother, heart disease and nervous troubles 
1 mother, heart disease and goiter 
1 mother, hay fever and asthma 

LONGEVITY OF GRANDPARENTS OF GIFTED 

It was thought that longevity of ancestors might throw 
some light on the virility of the stock from which the gifted 
children come. As the parents are in the large majority of 
cases still living, grandparents were used for comparison 
with the general population. Since not all of the grand- 



VITAL STATISTICS 129 

parents were dead, it was necessary to take account of the 
life expectation of those still living. The longevity index for 
grandfathers only was calculated. The formula used was 
supplied by Dr. Truman L. Kelley and is as follows : 

) in which 



N 

SLJxd Total life expectation according to U. S. Life Tables of liv- 
ing grandfathers of gifted at time of questionnaire, plus age 
at that time. 

2D*d = Sum of deceased grandfathers' ages at time of death. 

2LJxb = Total life expectation of grandfathers reckoned from the 
birth of the child's parent, plus age at that time, for grand- 
fathers who are living. 

SDJxb = Total life expectation of grandfathers at time of birth of 
child's parent, plus age at that time, for grandfathers who 
are dead.* 

* The meaning of the individual symbols is as follows : 

2 = summation 
L = living 
D - dead 

T = expectation of life at any given age, plus age at that time 
\ = date of questionnaire 
b = date of child's birth 

The comparative figures are taken from the 17. S. Life 
Tables, by Glover. Table 4 (for all white males in the 
original registration states) was used. 

Substituting the values as found for the above formula 
we have 

(33 007.265 + 42,907.5) - (27,454,48 + 45,827.148) 
___ l,H7 

Solving gives 2.35 years of life, in excess of expectation, 
for grandfathers of gifted children. 

This difference of 2.35 years is, for two reasons, too small. 
(1) It is based upon the theoretical expectation of life under 
the conditions prevailing in the United States in 1921. How- 
ever, many of the grandfathers died a number of years ago, 
a considerable proportion of them in foreign countries, and 
the hazards of living to which they succumbed were greater 
than those in the United States in 1921. Under modern Amer- 



130 TRAITS OF GIFTED CHILDREN 

lean conditions it is fair to assume that on the average they 
would have lived somewhat longer. (2) The formula gives 
credit only for excess years of life already completed. It 
allots to the grandparents still living only as many additional 
years of life as to the average man of equal age. These grand- 
fathers have already lived an average of 2.35 years of life in 
excess of the average man who lives to the age which the 
grandfathers had attained at the time of the birth of the 
gifted child's father, and the future is more likely to increase 
rather than decrease this advantage. 

TUBERCULOSIS AND LUES AMONG RELATIVES 

Under the caption "Tuberculosis" the physicians recorded 
the information given by mothers relative to history of tuber- 
culosis in parents and other near relatives. The following 
figures are therefore not limited to existing cases of the dis- 
ease, but are based upon all cases for which a history of tu- 
berculosis was reported. 

TABLE 32 
TUBERCULOSIS AMONG RELATIVES 

Dr. Moore Dr. Bronson Total 

% of % of % of 

Yes No subjects Yes No subjects Yes No subjects 
Tuberculosis in 
one or both 
parents 19 213 8.2% 8 262 3.0% 27 475 5.4% 

Tuberculosis in 

direct line 

(parents or 

grandparents) 41 191 17.7% 28 242 10.4% 69 433 13.7% 

Tuberculosis in 
family, includ- 
ing parents, 
grandparents, 
uncles, aunts, 
and cousins 73 159 31.5% 49 221 18.1% 122 380 24.7% 

The per cents in Table 32 are per cents of children, 
not of relatives. The figures reported by Dr. Moore (Los 
Angeles) are in each case much higher than those reported 
by Dr. Bronson (San Francisco Bay region). This is prob- 
ably accounted for by the fact that the climate of Southern 
California attracts many families in which there is a ten- 



VITAL STATISTICS 131 

dency to tuberculosis. The reports for parents are probably 
fairly accurate and show that about one child in twenty has 
at least one parent with a history of tuberculosis. The reports 
for more distant relatives are doubtless less complete. From 
these figures one could hardly infer that the incidence of 
tuberculosis in these families is excessive. 

The data on lues (syphilis) can not be taken as complete. 
The reports include- only two cases, one a father, one a great- 
uncle. There are two other probable cases, both parents of 
one of the children. 

HEREDITARY DISEASES AND DEFECTS 

Histories of insanity among relatives were reported to the 
physicians as follows : 

Pe^ cent of 
Male Female Total Gifted Group 

Parents 224 0.4% 

Grandparents and 
great-grandparents 11 5 16 0.3% 

Other relatives, sex not 

always stated 37 

Six families report 2 cases each; one, 3 cases; and one, 6 
cases. Thus, 1.6 per cent of the families furnish 36.8 per cent 
of all cases of insanity reported. The information regarding 
the last-mentioned family was not reported on the medical 
blank, but was furnished by the superintendent of the or- 
phanage in which the child is living. The family of this child, 
on the maternal side, is reported to consist almost entirely of 
insane, criminal, or mental defectives, although one cousin 
is reported as "very brilliant and a talented violinist." We 
have no reliable information as to the paternal ancestry of 
this child. 

Only 15 cases of feeble-mindedness among relatives were 
reported to the physicians: none in parents, grandparents, or 
great-grandparents; one sib of a gifted child; other relatives, 
14. This does not include the information just mentioned 
which was furnished by the orphanage superintendent for 
one of our group. One family reports 5 cases (four uncles 
and one cousin, all in the paternal line) . This is a third of the 
total number of cases reported. 



132 TRAITS OF GIFTED CHILDREN 

There were 19 cases of epilepsy reported: direct line, 
none; sib of child, 1; grandparents and great-grandparents, 
6; other relatives, 12. 

Information regarding "other nervous" tendencies was 
regularly asked for by the physicians. Under this heading are 
recorded hysteria, insomnia, and cases described as "very 
nervous," "high strung," etc. For the most part the defects 
mentioned appear to be of relatively minor importance. The 
following cases were reported: sib of child, 1; fathers, 
10; mothers, 24; other relatives, 30. Of the 65 cases, 54 were 
reported by Dr. Bronson. Dr. Bronson's figures are high 
largely for the reason that she included in her positive rec- 
ords the mothers who impressed her during the interview as 
having nervous tendencies, whether they reported such ten- 
dencies or not. 

Asthma and hay fever, which are frequently classified as 
nervous diseases, were reported as follows: 

Per cent 
Male Female Total of Group 

Asthma; 

Gifted Children 426 1.02% 

Sibs of Gifted 325 

Parents of Gifted 9 7 16 1.6% 

Other Relatives 15 

Hay Fever: 

Gifted Children 202 0.3% 

Parents of Gifted 13 11 24 2.4% 

Other Relatives 50 

Goitre: 

Eighteen cases were reported. 

Gifted Children 1 

Sibs of Gifted 2 

Parents of Gifted 9 

Other Relatives 6 

Total 18 

Cancer: 

Fifty-two cases of cancer were reported, 38.5 per cent of which 
were reported by 1.2 per cent of the families. 

Parents of Gifted 4 

Grandparents and Great-grandparents 38 
Other Relatives 10 

Total 52 



VITAL STATISTICS 133 

Only three cases of chorea were reported in the medical 
blanks; one of the gifted children, one sib, and one parent. 
The Home Information Blank reports one additional case 
among the gifted children. 

One father is reported as having palsy, one mother as 
having severe hysteria, with what appears to be hysteria 
hemiplegia. These are not included in the data mentioned 
above on "other nervous" tendencies. 

If the figures on insanity, feeble-mindedness, epilepsy, 
and other nervous diseases are taken at their face value, the 
showing made by these families is much superior to that 
which would be found for a random group of families in the 
general population. However, it is unsafe to assume that 
these mothers reported all cases of nervous defect of which 
they had knowledge, notwithstanding the assurance given 
them that the records would be kept in locked files and 
treated as entirely confidential as far as individual cases 
were concerned. In view of the possible error from this 
source, it is hardly worth while to make any comparison of 
the findings with statistics for the general population. 

SUMMARY 

1. The 91 mothers who are 45 years old or older report 
an average of 3.40 births, exclusive of miscarriages. When 
allowance is made for the greater chance of large families 
being caught in our survey, for childless marriages, and for 
celibacy, the average number of offspring per individual in 
the stratum represented by the parents of our subjects is 
0.72. This is 50 per cent lower than that found for the pre- 
ceding generation and is far too low to maintain the stock. 

2. There is a low but probably significant negative corre- 
lation (~.214:.07) between schooling of mid-parent and 
number of living births. There is a correlation of -.271 
.062 between IQ of child and number of living births. The 
latter correlation is only reduced to -.246 when the effect of 
schooling is eliminated by the method of partial correlation. 

3. The proportion of miscarriages to pregnancies in the 
completed families is probably not more than half the nor- 
mal for mothers of the generality. Infant mortality in these 
families has also been extremely low. 



134 TRAITS OF GIFTED CHILDREN 

4. The average age of the father at the birth of a gifted 
child was 33.63 years, (S.D., 7.70) ; of mothers, 29.01 years, 
(S.D., 5.64). These figures are slightly lower than those re- 
ported by Cattell for parents of American men of science. 

5. The data on order of birth, as far as they may be con- 
sidered valid, are in striking agreement with CattelFs figures 
in showing a preponderance of first-born gifted in families 
of two or more. The fact that superiority of the first born 
registers in childhood as clearly as in the achievements of 
adult life suggests that the causes are to be sought in native 
endowment rather than in environment and education. 

6. The number of deceased fathers reported is 42; of de- 
ceased mothers, 13. The causes of death are too scattered to 
be statistically significant. 

7. Forty fathers (8 per cent) and 61 mothers (12 per cent) 
have one or more chronic illnesses, many of which are of 
minor importance. This is probably less than would be 
found for adults of corresponding age in the general popula- 
tion. There is a record of tuberculosis in one or both parents 
of 5.4 per cent of the children. 

8. The longevity of the grandfathers of the gifted sub- 
jects is at least 2.35 years, and probably more than that 
amount, in excess of the expected. 

9. Four parents (0.4 per cent) , 16 grandparents and great- 
grandparents (0.3 per cent) and 37 other relatives were re- 
ported to the physicians as having had a record of insanity. 
Very few other cases of hereditary defect were reported. 



CHAPTER VII 

ANTHROPOMETRIC MEASUREMENTS 
Bird r. Baldwin 

In the spring of 1923 Dr. Terman invited the writer to co- 
operate in a comprehensive investigation of a group of 
intellectually gifted children whom he had selected in Cali- 
fornia. The aim, scope, and methods of the entire investiga- 
tion are described in the previous sections of this book. The 
problem assigned was to make an anthropometric study of 
the status of the physical growth of these children. 1 The pur- 
pose was (1) to secure an accurate picture of the status of 
physical development of each of the gifted children through 
a series of selected physical measurements; (2) to make a 
comparison of their total and partial growth with that of 
other groups; (3) to determine the relationships of various 
physical traits measured; and (4) to analyze the correspond- 
ence of the physical status and mental status of this group 
of children. 

METHOD OF WORK 

In order to insure good anthropometric laboratory condi- 
tions, it was decided to establish centers equipped with 
standard laboratory facilities at Los Angeles, San Francisco, 
and Stanford University for the three main districts. The 
children from the suburbs of Los Angeles and the surround- 
ing small towns came to Los Angeles to be measured, those 
from Berkeley, Alameda, and Oakland, to San Francisco, 
and those from the towns near Palo Alto, to Stanford Uni- 
versity. Many of the children lived so far from the center 
that an entire day was required for the parent and the child 

x Miss Beth Wellman, research assistant in anthropometry and psy- 
chology in the Iowa Child Welfare Research Station, cooperated in all 
phases of this investigation and made all the measurements under the 
writer's immediate supervision. The measurements were recorded by Miss 
Marguerite Drew, a graduate student in Stanford University. Material help 
has been given by other assistants in the Station. A grant of $1,000 in ad- 
dition to Miss Drew's salary was allowed from Doctor Terman's budget 
for the expenses of securing the original data. 

135 



136 TRAITS OF GIFTED CHILDREN 

to make the trip. At each of the centers the laboratory was 
located in the most accessible part of the city; it consisted of 
a measuring room, waiting room, and small dressing rooms. 

The parents were requested by a form letter to bring their 
children at specified hours on days designated. Arrange- 
ments were made with the superintendents of schools in the 
various cities for excusing the children from school work 
whenever it was necessary. Almost all the children notified 
came to the laboratory. 

The measuring began the first week of April and continued 
for thirteen consecutive weeks. The total number of children 
measured was 623, most of whom belonged to the group 
which has been designated as the main experimental group. 
Twenty-nine of these were not included in the statistical 
treatment because they had IQ's of less than 130 (although 
they had special abilities) or because they had been exam- 
ined by group intelligence tests only. The 594 children in- 
cluded 312 boys and 282 girls, with IQ's ranging from 130 to 
189. They were distributed by age as follows: 

Age in Years 1 Boys Girls 

2 1 

3 1 
422 
535 
664 

7 10 13 

8 14 15 

9 29 23 

10 35 42 

11 68 48 

12 63 45 

13 50 46 

14 26 29 

15 5 9 

312 282 

Measurements. Thirty-seven careful anthropometric 
measurements were taken on each child, without clothing, 
making a total of 21,978 measurements on the 594 children. 
The original records giving the data of the measurements for 

J In this chapter age is taken to the nearest birthday. For example, 
age 10 means 9 years, 6 months, to 10 years, 5 months, and 29 days. 
The above ages are as of date when the physical measurements, were taken. 



ANTHROPOMETRIG MEASUREMENTS 137 

each child are on file at the Iowa Child Welfare Research 
Station and in duplicate at Stanford University. The meas- 
urements selected were : 

Height: 
Standing 
Sitting 

Stem length to vertex 
Stem length to sternal notch 

Head: 

Anterior-posterior diameter 

Transverse diameter 

Height 

Circumference 

Width of face 

Length of face 

Shoulders : 
Width 

Arms: 
Span 

Length from shoulder to elbow, right and left 
Length from elhow to finger tip, right and left 
Circumference of upper, relaxed and flexed, right and left 
Width of wrist, proximal and distal, right and left 
Circumference of wrist, right and left 

Chest: 
Width 
Depth 
Circumference 

Hips: 

Width at ischia 
Width at trochanters 
Circumference 

Legs: 

Length 

Breathing capacity 

Grip: 
Right 
Left 

Weight, nude 

All measurements were taken in the metric units and the 
final results translated into the English units for this report. 
The actual time required for making the measurements for 
each child with the aid of the recorder was about 20 minutes. 

Figure 2 gives the location and actual measurements of 



138 



TRAITS OF GIFTED CHILDREN 



31 physical traits on the fourth boy measured. He is 6 years 
of age, and has an intelligence quotient of 154. He holds a re- 
markable record for swimming and diving. He also plays the 
violin for public exhibition and is a movie actor. The boy is 
slightly below medium height and his weight is slightly 
above the Iowa pre-school norm for his height. He is above 

FIGURE 2 
LOCATION OF PHYSICAL MEASUREMENTS 




The location of the measurements made on all the mentally superior California 
children are shown by the lines in this diagram. The vertical lines to the right 
indicate the height measurements; the ellipses, the circumferences; the short 
horizontal lines, the diameters; the long horizontal line through the shoulders, 
the span of arms; and the slanting lines on the head and chest, the depths. 



ANTHROPOMETRIC MEASUREMENTS 139 

the norm in breathing capacity and chest measurements, 
both for his height and age and for his weight and age. His 
shoulders are broader than those of the average child of this 
age. 

Instruments. The apparatus and instruments were those 
used in the Iowa Research Station Laboratory, including the 
measuring board for height, benches, square, plumb-line, 
sliding and spreading calipers, tapes, and metric rule, supple- 
mented by scales, a spirometer, and a hand dynamometer 
from the psychological laboratory of Stanford University. 
All instruments except the hand dynamometer were care- 
fully standardized before the work was started; the hand 
dynamometer, Smedley No. 256, was later sent to the Bureau 
of Standards, Washington, where corrections were found nec- 
essary after careful testing. Since the errors ranged from 
-2.8 kg. to -1.7 kg. at five unit intervals up to 50 kg. on the 
scale, the average error was found to be -2.17 kg. for the 
boys and -2.33 kg. for the girls. The averages, deviations, 
probable errors, and coefficients of variability given for grip 
are for the corrected measurements. 

The technic of taking the measurements was that de- 
scribed by the writer in his Physical Growth of Children 
from Birth to Maturity, 1 and revised in his new book on the 
same subject now in preparation. 

Cooperation of parents and children. The attitude and 
the cooperation of the parents and children were exception- 
ally good throughout the entire study. In general, the chil- 
dren appeared to be physically well-developed and normal. 
Mentally they were alert and quick to respond. Socially they 
were well-mannered and showed good spirit. The parents of 
the mentally superior children, as a rule, showed a great in- 
terest in the welfare and training of their children. In a few 
instances we were able to get valuable data on the physical 
growth of the children from original diaries. Figure 3 gives 
the individual growth curves from birth of a brother and 
sister included in the study. The girl is of medium height; at 
6 years 6 months, she is 0.7 per cent underweight, average in 
breathing capacity, 1.1 inches above the norm in width of 

^Baldwin, B. T.: Physical Growth of Children from Birth to Maturity. 
State University of Iowa, Iowa City. Studies in Child Welfare, 1921, I, 
No. 1. p. 411. 



140 



TRAITS OF GIFTED CHILDREN 



shoulders, and 0.9 pounds below the norm in right grip. The 
boy is also of medium height; at 11 years 2 months, he is 8.7 
per cent overweight, 10 cubic inches above the norm in 
breathing capacity, 0.1 inch above the norm in width of 
shoulders, and 8.0 pounds below the norm in right grip. The 

FIGURE 3 

INDIVIDUAL GROWTH CURVES OF Two SIBS, BROTHER AND SISTER 
50 




40 

in Months 

The boy underwent tonsillectomy at 39 months, and had chicken-pox at 52 
months, and whooping cough at 57 months. The girl had whooping cough at 2 
months, double bronchitis at 6 months, diphtheria at 35 months, measles at 37 
months, removal of tonsils and adenoids at 62 months, and bronchitis at 65 
months. The effect of the diseases is seen in the irregularity of gain in weight of 
both sibs. 



ANTHROPOMETRIC MEASUREMENTS 141 

curves show steady growth in height, but somewhat retarded 
growth in chest development and irregular growth in weight. 
They also show that the diseases indicated, except diphtheria 
for the girl, affected growth in weight, but had little appar- 
ent effect on height or chest growth. There is a drop in weight 
following measles for the boy, followed by marked gain. For 
both the boy and girl there is an acceleration in weight after 
the removal of tonsils and adenoids. Since reclining height 
is longer than standing height for the same individual, the 
height curves dropped back when standing height was first 
taken instead of reclining height at 19 months for the boy 
and at 14 months for the girl. 

When the work on physical measurements was begun, 
the parents had already acquired an intelligent interest in 
the larger survey of the characteristics of mentally superior 
children and had become acquainted with the results of spe- 
cialized examinations. In order to meet the very large num- 
ber of requests for the results of the physical measurements, 
a form letter was forwarded to the parents or guardians for 
each child. The letter is reproduced on pages 142-143. 



.42 



TRAITS OF GIFTED CHILDREN 



C/2 

H 

W 



W 







CO 

H 



53 



8 



O 
P, 
W 



Cfl 

t-( 
0) 

'3 
p 

-d 

I 

d 

ea 

55 



o 

o 

"S 



GJ 

I 

d 



OS 



o 

tlD 

fi 
ctf 
43 
O 

>> 

cs 



CO 
OJ 



H4 

Ll 

s 



.2S 
"2J 5 

sb 

O 

tn 
O 



d 
o 

a 

o> 
S 

CO 

cs 

0> 

a 



cd 
Q 

CO 

O 

a 



^ 



3 

CO 

ctf 
o> 



A 



2 
2 

o 

*s 

o 



o 

2 

tc 



o 

ed 



" 




B 
s 



C/3 
CS 



2 

o 







o . 

CJ co 

M TJ 
L< 
<S 

T3 

<rt 
*- 

CO 

d 
; rt 

& 

H *t- 

o S 



fi cc -M ' 



H if A S 2 a *S 

W'^r^ Q 

"iS co P 1 ;i S *-> - >^ 
6 *^ - 1 



M 



*" 



J 72 ^ O 



CO J 

O .. 
^Q oo 



O O 



-< ^T O . C JW 

^ S * "JjJ "3 



>> S, 



0^3 O*O rd 

*.. ^ _ ^ aT p 



fl Q) L< CO 43 T3 L< 

s ^a g -|H *^ fl *g 

O C( rQ ^a . OS r< 
C5 '^ 4- r-1 r? ^H 






tO 

1: 







, 

X S g co g.SP^ 

l 



s 



! L< ^ p 

!<S Q 



S .0 



V-t CSS 

<i> 



.2 ti 



g'S 
o o 





ANTHROPOMETRIC MEASUREMENTS 



143 




S 



r-t 

al 






,d * rt j? 


,373 ** j? 




o 


4~* 5 ._f f 


"*" J C4 n ( P /T< 






4- 03 HH 13 






<tf 




O c3 H 




o 
<+> 


g w -S 


* ^ * J 5 




asure 


2^0^ 


a g^l 

^ S <s ? 

ui cp ,3 




o 

a 


o o cc a ^ 

.PH 0) 03 


a2 p< 
SS ^d 




CO 


^i bD 


CU O fl CD 




3 


rd J-< j.* C3 


*- 4- 
4- Ctf CS 




H 


.3*j g 


. u S ^ 
T) o S ^j 







CO ^ <T3 

^ ^ S.J5 




CO* 


o> 
,3 
o 


l^lg 


ItsiS 


(H 
O 


d 


O 4-> 

o * 


tJ "5 ^ ^ 


>4 


!-( 

CO 

O 
bn 


| C .^S1 

^ili 
-gggi 

O -^3 p5 5* 
0) +3 Oj ft 5 


5 d "^ "3 

|g|| 

T3 ^T O 

S'S d ^ 

co +2 2 oS 


sincerely 


for this a{ 
obustness. 


S oS 

|*agi 

si:N 

.In TJ SI tO 


^ ^ >> 

d ^ -H Q 


& 

(D 
> 


<O f-< 

^^ 

1 


o ^'53 8.S 

|S-2S| 


il g l 

2 ^S-S 




*& 


t^ S fa u o 


* S- 2 r 

^ cts rj t-i 




OLDERS: inches. The 
icates a child's general pi 


ve measurements give a c 
5 made. However, the me 
ven age, since children vt 

t tO knOW WHETHER THE 

child year by year and c 


3veral thousand individua 
shows that tall children 
although there are excep 
al stature earlier than sho 




0^3 

*f 

fe d 
o ^ 

wg 

S s 

3 


NOTE: The abo 
time they were 
child is at a gi 
more importan 
measuring the 


Our study of s< 
same children 
remain short, 
reach their fin 





-w 

g 



^ 

o 

H-a 
O 

S 

*> 



144 TRAITS OF GIFTED CHILDREN 

COMPARISON OF THE GROUP WITH OTHER GROUPS IN VARIOUS 
SECTIONS OF THE UNITED STATES 

Since the weight-height relationship of a child furnishes 
one of the best general criteria for its physical status, the 
average heights and weights for all the children included 
between the ages of 7 and 15 were first computed and the re- 
sults compared with those of earlier writers on California 
children and on a few representative groups in other parts 
of the United States. The results given in Table 33 show 
at a glance that this group, measured by the group average, 
is physically superior in both height and weight for age, al- 
though several children are small and some are considerably 
underweight. The Oakland children measured by Barnes 1 in 
1892 are considerably inferior to this group, although they 
were heavier and taller than similar groups of children from 
Boston, Worcester, Toronto, St. Louis, and Milwaukee, whose 
records were displayed at the World's Fair in Chicago in 
1893. The Oakland children later studied by Boas 2 were su- 
perior to those studied by Barnes, but inferior to those in- 
cluded in this study. The Davenport group represents a se- 
lection from the best residential district in the city. The Oak 
Park group is from one of the most favored social sections of 
Chicago. FaberV study, in 1923, was of a group of California 
children. The California gifted children excel them all in 
height and weight, for all ages included. They also excel the 
early Boas-Burk 4 averages for the country at large, when 
approximately 90,000 children for height averages and 68,000 
for weight averages between the ages of 5% and IS 1 /^ years 
were included. 

According to the standards in the new Baldwin-Wood 
tables for tall, medium, and short American children, based 
upon records from 124,000 well-developed American-born 
children measured without clothing, 44 of the 312 superior 
boys are tall, 233 medium, and 35 short. Of the 282 girls, 45 
are tall, 208 medium, and 29 short. Of the boys, 176 are of 
normal weight for height, 89 overweight, and 42 underweight. 

Barnes, E. : Physical Development of Oakland Children. Oakland School 
Rep. 1892-1893, XCIII, 38-44, 

"Boas, F.: On the Growth of First-born Children. Science, n.s. 1895, I, 
402-404. 

Faber, H. K.: Personal Communication. 

*Burk, F.: Growth of Children in Height and Weight. Amer. Jour/u 
PsychoL> 1898, IX, 253-326. 



ANTHROPOMETRIC MEASUREMENTS 



145 



tH 
O 



CO 

A 

I 



a 

O 

8 g 

< 



o 

fo 
o 



o 

3 

H 

w 

2 
w 

B 



I 



i 



>rj C3 

2 ^ 

W 5 



a 



O_ 

lli 



O O^ CO CO O^ 



COl^CirH COOJrHCOrH 
rHt^CNrH XO CO rH CO rH 
1Q ifj CO I s * t"* CO Oi O rH 



IVCOCSJO OOCOrHrHl 



CO CNJ 10 00 rH 00. 1O O O 



{Nt^iOO500 
OOOOOSO^H 



CO 
JO 



D CO rH ^ CO 00 
* Trf" CO 00 rH ^ 



XOCOCOt^ CQOSCTJC 
-- " coivoic 

XQXO1QC 



CO CM 00 
* 



*<^ CO O CO 00 

' 



COCO^rjTP OCOrHpt^ 

'00305 06 co i> xc o 
irSCOCO t^-OOOOrH 



cot-jooo. co<N<Ncqt>. 

T^iOCOCO t>COO5OrH 



rHCNICOCO OjOiCDrHOi 

co' c4 co co os od d od oo' 

Tt^lQiOCO CO t^ O5 OS O 



O(NI>OOO 



Oi CCJ O5 t*^| O C*^ ^ t^j CO 



CO CO CO CO I s -; rH t> C^ rH 



COOirH COCOOrHrH 
Tt^ ^ XQ IO UD CO CO CO 



XOOCOCO lOOrHCOO 
I s " O O-J CO CO 00 rH (M Xt 1 

^IOXQXO xoxocococo 



rHCNJCO^XO 



ng. 



tc 

a beg 



M 2 2 .S 

:l^|5! 



J5.2l?2 

i o rt T5 O 



hou 
lot 
thi 



^ W OjS^rS 
^S^ISl 

^fl-s ^ . : 

/5 2 

0^ jf rcJ r-, TH 

fl^tCg'd^ 
TJtH * d g 
,> fl^CD * 
.H* fl^HVO^ 

oto o^ M 

W H CO W O 

s,^s,ss 

S o a MM 

rtj tfi CO CJ ^ 

5 * a^ 

* <u S^pq^ 

8 fl 5l 
^ 5 l|s 

1^1^ 

^Qi ^ O 

Jilii 

32 ^^^^ 

!p-s 

rH o .i-j eft co . 

Sco-S^gS 

I "3 ^ I "^ 

i> i I ^ 



fornia 
n 



ifS^i'l 

4fSH 

-J'SaSJS 

alls^s 

o^ . rt 
2-0 2 ^ 
'M ^"5 6" 

rrt PJ wtrl >C 

i-M ^-i Oi P-4 ^ pQ 

cj J o J oi o* 

PQ WfQPQpqfe 



146 TRAITS OF GIFTED CHILDREN 

Of the girls, 143 are of normal weight for height, 88 over- 
weight, and 51 underweight. The criteria for overweight and 
underweight were a deviation from the norm of more than 6 
per cent for children less than 10 years of age, and 8 per cent 
for children 10 years and older. A more detailed analysis of 
tall, medium, and short children is given in Table 34. 

TABLE 34 
RELATION OF WEIGHT TO HEIGHT, GIFTED CHILDREN 





Normal 


Over- 


Under- 




weight 


weight 


weight 


Boys 








Tall 


52.3% 


31.8% 


15.9% 


Medium 


56.5% 


29.3% 


14.0% 


Short 


68.4% 


22.9% 


8.4% 


Girls 








Tall 


46.7% 


22.2% 


22.1% 


Medium 


51.9% 


33.2% 


14.9% 


Short 


48.3% 


31.0% 


20.7% 



DEVIATIONS FROM NORMS IN WEIGHT AND BREATHING CAPACITY 

In order to get a graphic representation of the deviations 
from a standard group in weight and in breathing capacity 
of the gifted children, the distribution of the percentages of 
deviation for each child has been charted (Figs. 4, 5, and 6) . 
The deviation in weight was calculated from the Baldwin- 
Wood weight-height-age norms for nude children, the devia- 
tion in breathing capacity from the unpublished Baldwin 
height-breathing capacity-age norms and nude weight- 
breathing capacity-age norms for American-born children. 

The curves for weight deviations show the range of de- 
viations from the norms for the boys to be from 24 to +60 
per cent and for the girls from 24 to +48 per cent. It should 
be noted, however, that 74.5 per cent of the boys and 62.4 
per cent of the girls deviate less than 10 per cent from the 
norms. 

One boy deviated 60 per cent below the breathing capac- 
ity norm, another 60 per cent above norm, but 72*6 per 
cent of the boys deviated not more than 10 per cent above or 
below the norm. One girl deviated 45 per cent below the 
norm and another 45 per cent above, but 68.3 per cent de- 
viated not more than 10 per cent above or below the norm. 

The results show a wide range of distribution on both 



ANTHROPOMETRIC MEASUREMENTS 



147 



FIGURE 4 

PERCENTAGE DEVIATIONS OF GIFTED CHILDREN FROM THE BALDWIN- 
WOOD WEIGHT-HEIGHT-AGE NORMS 
30 




tf 



zo 

of Deviation 

Distribution curve of deviations from the norm in -weight for height. The 
horizontal direction shows the percentage of deviation above or below the norm 
and the height of the curve shows the percentage of the children included for each 
unit of deviation from the norm. 

FIGURE 5 
PERCENTAGE DEVIATIONS OF GIFTED CHILDREN FROM THE NORMS* 

IN BREATHING CAPACITY FOR AGE AND HEIGHT 
30r 




-50 ^40 -30 -20 -10 
R 

1 Baldwin, unpublished data 



O 10 SO 

of Deviatfort 



30 



4O 



148 



TRAITS OF GIFTED CHILDREN 



FIGURE 6 

PERCENTAGE DEVIATIONS OF GIFTED CHILDREN FROM THE NORMS* 
IN BREATHING CAPACITY FOR AGE AND WEIGHT 




-60 



* Baldwin, unpublished data. 

the positive and negative side of the norm for both height 
and breathing capacity. For weight the results indicate a 
fairly normal distribution, with a slight preponderance to- 
ward the positive side of the curve. For breathing capacity, 
when compared with weight-height-age norms, the curves 
are skewed toward the positive side of the distribution. As 
a group these mentally gifted children are superior in 
breathing capacity to unselected children. 

RESULTS OF PHYSICAL MEASUREMENTS 

In Tables 35 and 36 the averages, standard deviations 
with their probable errors, and coefficients of variabil- 
ity for the measurements on the 37 physical traits are given 
for boys and girls from 7 to 15 years inclusive. The 24 chil- 
dren less than 7 years of age are distributed from 2 to 6 years 
of age and are therefore too few to treat by age groups. The 
average represents the central tendency of the group, but, of 
course, does not show individual variations or differences. 
The standard deviation is the amount of variation above and 
below the average in terms of the measurement for about 68 
per cent of the children. The coefficient of variability is a 



ANTHROPOMETRIG MEASUREMENTS 149 

percentage in terms of a pure number, whereas the standard 
deviation involves various units of measurement. In the case 
of height at 7 years, for example, the average for the boys is 
48.5 inches, while the standard deviation shows that about 68 
per cent of cases vary not more than 2.4 inches on each side 
of this average. The coefficient of variability in height for 
the 7 year old boys is 5 per cent. 

Since all measurements in Tables 35 and 36 were ac- 
curately taken by a trained examiner under standardized 
conditions, and since the number of traits measured includes 
a comprehensive and significant selection, the results estab- 
lish for the first time a series of averages for thirty-seven 
traits for a considerable number of mentally superior boys 
and girls. The less accurate measurements from one stand- 
point of technic of measuring are width of face, circumfer- 
ences of arms, and circumference of chest. 

Sex and age differences. A general survey of Tables 35 
and 36 and of Figures 7 and 8 shows interesting and sig- 
nificant differences in sex development. The boys are supe- 
rior to the girls in all traits up to 12 years of age, except width 
of hips, length of leg, and weight, in which the girls begin to 
show superiority before this age. It must be remembered, 
however, that these are averages of the measurements and 
that chronological age is an arbitrary division that does not 
take adequately into consideration the accelerated and re- 
tarded stages of anatomical and physiological ages for par- 
ticular children. The writer's individual growth studies on 
large numbers of children show that tall children are farther 
advanced than short children in their periods of growth dur- 
ing childhood. "This basic principle is also illustrated in a 
later section of this study (pages 163 ff). 

When 12 years of age is used as an arbitrary division, it is 
found that during the subsequent adolescent period the girls 
equal or excel the boys in standing height, sitting height, 
stem length to vertex, stem length to sternal notch, chest 
measurements, hip measurements, and weight. They main- 
tain their superiority for the ages included. 

Increments of growth. Figures 7 and 8 and Tables 35 
and 36 show an appreciable increase in increment of growth 
from age to age in practically all trait^, except head meas- 
urements, with an adolescent acceleration beginning at 
about 13 years for the boys and 11 years for the girls, the 



150 TRAITS OF GIFTED CHILDREN 

FIGURE 7 
AGE CURVES FOR SIXTEEN PHYSICAL TRAITS GIFTED BOYS 




ANTHROPOMETRIC MEASUREMENTS 

FIGURE 8 
AGE CURVES FOR SIXTEEN PHYSICAL TRAITS GIFTED GIRLS 



151 




10 



152 TRAITS OF GIFTED CHILDREN 

acceleration varying with different traits. For instance, the 
acceleration in weight precedes the acceleration in height. 
The ages do not extend far enough to make possible a deter- 
mination of the final status of growth in any one of the traits 
included. 

Variability of measurements. A study of the lineal meas- 
urements, taken in terms of inches and tenths, shows that 
the least variable are those of the head (excluding the face), 
standing height, sitting height, stem length, width of shoul- 
ders, and width and depth of chest. The most variable lineal 
measurements are those of circumferences of the wrist, chest, 
and arms. Weight is from 2 to 3 times as variable as the 
lineal traits, and the psychophysical traits of breathing 
capacity and right and left grip show considerably more 
variability than weight. 

A comparison of the coefficients of variability of all traits 
for all ages for the two sexes shows that, as a rule, the girls 
are more variable than the boys, and that the variability in- 
creases during the adolescent acceleration in growth. The 
percentage of variability differs for the sexes with the stages 
of physiological growth attained. 

Indices of growth. The percentage of the weight to the 
height, that is, the weight-height index, varies with age and 
shows that the girls begin to increase in weight for their 
height after 12 years of age, and the boys slightly later. The 
individual indices vary with the relative increase in height, 
which also varies at a particular age with the height status 
of the child for each sex. The majority of these children are 
of normal weight for height, but several are decidedly under- 
weight and others equally overweight. 

The average cephalic indices are approximately the same 
for boys and girls and vary little with age. 

The cranial capacity was computed for the children for 
ages 7 to 15 by the Pearson-Lee formula for general compari- 
son with norms of Porteus 1 and others for Anglo-Saxon boys 
and girls. The results for the gifted children are above the 
Porteus norms throughout. The tabulated results are not in- 
cluded here because, as Professor Pearson writes in a recent 
letter, "to assume the formulae hold for children, when the 
bone-growth (both internal and external) is so very consid- 



, R. J. H., and Porteus, $. D.: Intelligence and Social Valuations. 
Res. PubL No. 20. Training School, Vineland, May, 1920. 



ANTHROPOMETRIC MEASUREMENTS 153 

erable during the adolescent period, might lead to grave 
error." (Cranial capacities are now being computed from a 
larger series of head measurements on various children in 
the Iowa Child Welfare Research Station laboratory under 
Professor Pearson's directions.) 

The average chest index varies little with sex or with age, 
except at 15 years of age, when the girls exceed the boys. 

On the average, the sitting height is approximately 53 per 
cent of the standing height for boys and for girls of the ages 
included. Arm span and height are approximately equal, 
since the difference is less than 1 inch for both boys and 
girls at any age. 

Comparison with control group. When a comparison is 
made with the Oak Park children, with whom the same 
instruments and methods of measuring were used, from the 
ages of 7 to 14 years in arm span, width of shoulders, width 
of hips, and right grip, it is evident that the gifted boys and 
girls as a group are superior in these four traits. 

COEFFICIENTS OF CORRELATION FOR PHYSICAL TRAITS 

Zero order correlation. Table 37 gives the coefficients of 
correlation by the Pearson formula between four physical 
traits for the ages 9 to 14 inclusive and six additional physi- 
cal traits for age 11, for boys and girls. The results show 
relatively high positive correlation between weight and 
standing height for all ages for both sexes, the correlation 
for the boys being higher than that for the girls. 

The coefficients of correlation between breathing capacity 
and grip are also positive and relatively high for all ages for 
boys and girls, which means that for children of the same 
age, those of superior breathing capacity are also of superior 
strength, as indicated by grip of the right hand. At 11 years 
the coefficients of correlation between breathing capacity 
and width, depth, and circumference of chest are positive for 
boys and girls, but higher for boys than for girls. The results 
indicate that width of chest has a higher correlation with 
breathing capacity than either depth or circumference of 
chest. 

These correlations indicate that for the various ages in- 
cluded the children who are relatively high in one physical 
trait are relatively high in the other physical traits. 



154 TRAITS OF GIFTED CHILDREN 



00 sH I- 

CO OO O> 

3 i 5 I i | 111 1 5 

t> " ' 10 oo 

^ o^ oo 



K 

03 CM 

O CO 

H) ,-v O 



fe 
o 



+ ^ ^s +1 

CO ' ' ' ' Jj <3 CM 

S3 3 



"*^ 10 f** 

i f 1 H 1 I I ' I ' I H 

' CO *"< 

Sp, CO *O 

^ J CO OOCMiO tO * i> CO O -^ "^ ^ "* CO 

-^ftCOCM -^ CO CO CO ^ ^lOCO I s * COcOOO t*^cO 

2 ^ N o O O OOO O^ ^-soOO O OoO <^O 

fi cs S Jg 4- f +J +1 +1 +1 +1 -hi +t 5 +1 + ' + ' + ' *' + ' +l +l + ' 

IH & ^-s T HOi-rH T-<'^t t '^ T-Kt- 1 >^-' 1 (CO 1 ^f CO CMcOCM OicO 

PH |-*^ t^*. 1^3 10 Oi xO CO ^** CM *^ ^^ ^^ ^"* ^^ ^""^ C^ 

M QQ t> oo co "^ *rf< **< ^ !> CD oo 10 "^ "^ ^ "^ ^ 

CJ 
M 

^ -^i O CO t"* 

ss * s II ! ! l 'l '3 IH ' ! ' ' ' ' 'H 

F* oo "^"^ r^ co 

W 2 ^ ^. LO ^5 

>J < 

3 S ^ 

5 W C5 J^ O r ^ 

I - 5 1 i MI i+j S i i i in iTi 

I ^ CM %* ^ 10 Ig 

r. ^ 

CQ 

C Cflt/3l/3O S 501 ^*^^ 

a ri ii| a is f ii| a 

I .S? ? I ? -S ?o~ f 



^oooo^^i ^^ woo da 

6 ^ fe-23 3! ? a S-25 5 6 I 

g S2^32 S^S , S213| I S& -g 

2 1-s|.lf ft? T 1-llJ.f t fTT T 

g s-^-sai Ui i s-aj-ajl i Ui J 

W.j laD**-!**^^^,^^^..* ^-* ^. toD**-^**- 1 ^^..^ ^_t^-^^J 

*- ^{ r^ O o <D *ti -T^ -Tn ^S ?* i r-T O O ^ ^ U -< .FH .J3 

3 cd .5ii p S'S ^^ ^.Sliz^ "Si^ ooo 

w -^.d^^toCrtasas as ^rg.n^r-jjo-^ ... 

c*i*id*SftftPH ft ^c^SS*^ 

cS^SrSQ-sCSoSrt ctf ,j 5 S S 2 * 

-g -g ^ ra o o w o flSg'S^^ 



-d -i T3 *^ w ty w w .S -* -i < ^- l v^^3 wvw w 

j^^^toCrtosos as ^r-j^cl^H'-; 2P 'TJ cs cc as 

!*i' i :&i 1 S' S-^lsSl^I &&& " 

x M Jg>^-oou Ofl2>>^-2rC3 ooo ti 

a T TTJ? g5P5P1^5P 1 TTT^J2S2P?P- 



ANTHROPOMETRIC MEASUREMENTS 155 

Partial correlations. By use of the partial correlation, 
age was made constant for this entire group. There is still a 
positive correlation between the physical traits included. 
For the boys, the lowest partial coefficient of correlation is 
+.483, between breathing capacity and depth of chest; the 
highest is +.703, between standing height and weight. For 
the girls, the lowest partial coefficient of correlation, +.466, is 
between breathing capacity and depth of chest; the highest, 
+.824, between breathing capacity and circumference of 
chest. These results show that when the effect of age is 
eliminated 1 children who are superior in one physical trait 
are likely to be superior in other physical traits. 

Comparison with control group. For control in the study 
of the correlation of physical traits of these gifted children, 
data on Horace Mann 1 and Oak Park 2 school children were 
used (Table 39). The Horace Mann group included 120 
children for each age between 7 and 16 years, who had been 
measured without clothing consecutively each half year for 
periods of 6 to 10 years each. The Oak Park group included 
122 children for each of the ages 11 and 14 years, measured 
with clothing. The three groups are of practically the same 
social status. 

The coefficients of correlation between the various traits 
are as a rule higher for the Horace Mann children, who have 
had a long series of consecutive measurements. The corre- 
lations for the gifted children are about the same as those for 
the Oak Park children. The relative position of the three 
groups with respect to each other varies with the age, sex, 
and the trait measured. The low correlation between breath- 
ing capacity and right grip for the California children at 9 
years of age may be due to the fact that the California chil- 
dren had not had previous training or experience with these 
two psychophysical tests. 

This series of comparative coefficients of correlation 
shows that growth relationships for the traits included are 
very similar for the three groups of children. The boys of 
the three groups have as a rule higher coefficients of correla- 
tion than the girls. 

Baldwin, B. T.: Physical Growth of Children from Birth to Maturity. 
State University of Iowa, Iowa City. Studies in Child Welfare, 1921, I, 
117-139. 

2 Reference is to unpublished data. 



156 



TRAITS OF GIFTED CHILDREN 



o "H 



CO 
CO 



1^ 
CM 

VO 



1 Hi i 


CO *<f CO CD T 
<O l> OO <O Ci 
t** iO *^* CO C 


05 




fe 









*"j OO iO ^J* TH 

<"} j 1 _ j Al 


1^5 CO CO CD CO C*^ CO O 


T3 < ' T^ ir"4 CM 

* . 2 o <=> 

^ 3+1+1 +! +| 


oooooo oc 

H +r +i +i +1 +r +f + 


|4 ?S <^f( if) <^ J, 


t^CMC^COC^CO ^f 


"^5 O9 OS CD Cft 


O> CM CM CO CM CO. CM T 


g 00 I> 00 CO 


c^ooooooooc- ooo< 


S a 




M 




oZ 3 "Si 




fi H *2 








^ CJ T-( 00 


Ol>COg-^03 ^t<0 


o w o o o o 


ooo<o<oo <oc 


<! g +f +f +f +f 


+i +i +i +1 +i +F +f + 


|J CM *^ OO CM OO 


lO CM CM CO tH CO O T- 


3 ^ CM as T-t 


oo^^t^osr"* c^r 1 


o oo t** co co 


co t** ^ oo i> r* oo o 


O 




fc 




o 




K 




K 




CJ 


** 


^ 


2 


u 
H 





S s 

, to w 

co -s S S 

fe ^* >M ,M 

Q 44 o o 


o-3 - ^ S 

QJ CtJ W W o 

S 9 III 

S - o, ^ - 5 1 


S *** *o *S 


QJ J* 1 M< Q ,. ^4 . 

^ q *C S ^ 


^ f3 ^_i 


S "jg & & ^ 4. 


a "g "S 5 6 

1 1 | 8- 
u T^ f f * 


1 I 1 I 1 1 i 1 

! t f f K L 1 , 

CU <(j rfi i* ^_t 4.* -*- 


j^ tao to to tic 

^4 KJ 03 03 CQ 


9 I J. 'S 'S -3 *e 

. **1 flj cC cO C 


PC , < rH P-H . > i 
<< as aj cd as 

<2 , .2 . .2 

O O O O 


1 1 1 | s t g S 

to Q> t ^ toto totj 
^^S^^aa qc 


fli o o o o 

^ G fl fl (4 


g g .| 3 3 3 2 


5 o o o o 

g (N U h t+ 
o .^1 P] *JCJ JCl 

EH O 3 U 


OOpq'S'S "S"c 

S^bSs^^ <uc 
pMA^jdJ^tHt^ u t- 

UUUKWCQ Mff 



ANTHROPOMETRIC MEASUREMENTS 



157 



RE 



W 

o 

fe 
o 

CO 

PU 

1 

O 



W 

o 






I 

o 

o 

CO 



W U 
H W 





fc 

i 







02 
H 



COEFFIC 



l 

bag 

it 





t 








5 







p, 





XO 
t^ 





CO t 












<M 

CO 



TH 
) CO 



+1+1 
!> 1ft 



T-H 1O 

i i lo 



CD iH 

+i+i 



S 



1 



OO CO 
CO t" 



s 



X 

OO 



ol3 o 13 
WO Wo 



fl 

a 



^ ^J ^ *- a 

o 13 o 13 
WOO Wo 






-i- 



CO Xa 

*H rH 



C^ OO CO *CH ! 

CO CO CO t"-* 

^ 



+1+1 +I+I+I +1+1 +1+1 +I+I+I 



*sj 



+1+1 +1+1+1 +1+1 +!+! +I+I+I 

CM rH CO CD f* Ol t** O5 O<l CO XQ ^$i 
CO XO XO XO XO XO CO CO CO CO l> CO 



Xt) 



C3 

a 



r 

o o 15 13 
WO WOO 



158 TRAITS OF GIFTED CHILDREN 

RESEMBLANCE BETWEEN PARENTS AND CHILDREN 

When the parents brought the children to the laboratories 
for measurement, the height and weight of the parents were 
taken in order to determine the correlation between the 
height of parents and the height of their children. In a few 
instances the fathers were measured at their homes. Table 
40 gives the correlations of height of father and son and 
of mother and daughter for 305 children for the ages 10, 11, 
12, and 13. The average of the four correlations for fathers 
and sons is .36; for mothers and daughters, .305. 

TABLE 40 

COEFFICIENTS OF CORRELATION BETWEEN HEIGHT OF PARENTS AND 
HEIGHT OF 305 GIFTED CHILDREN 

Age 
10 11 12 13 

No. of sons 25 49 42 37 

r, fathers and sons .439 .109 .289 .088 .213 .099 .501 .083 

No. of daughters 36 40 38 38 

r, mothers and 
daughters .323 .101 .328 .095 .324 ,098 .245 .103 

RELATION OF PHYSICAL GROWTH TO NUTRITIONAL STATUS 

According to Dreyer's method 1 the normal weight of an 
individual is computed by finding the weight, in Dreyer's 
tables, corresponding to the subject's length of trunk and 
circumference of chest; the average of these two weights is 
the normal for the person of the given measurements. 

Weight derived from , Weight derived from 

length of trunk ' circumference of chest 

normal weight. 

2i 

The percentage below or above normal is readily computed 
by comparing the observed weight and the normal weight 
derived from the tables. 

(Observed weight ~ calculated weight) 100 

~ per cent deviation. 

calculated weight 

Greyer, G.: The Assessment of Physical Fitness. N. Y. Hoeber, 1921. 
p. 127. 



ANTHROPOMETRIC MEASUREMENTS 
Dreyer's standards of weight are : 



159 



Normal 

Possibly abnormal 
Probably abnormal 
Certainly abnormal 



within 5 per cent 

plus or minus 5 per cent 

plus or minus 10 per cent 

plus or minus 15 per cent 



An inspection of Table 41 shows that the Dreyer method 
of predicting weight on the basis of sitting height (stem 



TABLE 41 

DISTRIBUTION OF GIFTED CHILDREN ACCORDING TO DREYER'S STANDARD 

OF WEIGHT 



Boys 

Weight, Per Cent 
Above Normal 

15 or more 

10 to 14.9 

5 to 9.9 

to 4.9 

Below Normal 

to 4.9 

5 to 9.9 

10 to 14.9 

15 or more 



Age 

7 8 9 10 11 12 13 14 15 



19431 
1 2 6 13 11 8 7 
1 10 12 23 15 20 8 
4 10 8 12 19 14 5 



4366 10 8431 
12 113 12 



9 12 28 34 68 60 49 25 5 



All Ages 
Number Per cent 



18 
48 
90 

77 



45 
11 



290 



6.21 
16.55 
31.03 
26.55 



15.52 
3.79 

0.34 
99.99 



Girls 

Weight, Per Cent 
Above Normal 
15 or more 
10 to 14.9 
5 to 9.9 
Oto 4.9 

Below Normal 

Oto 4.9 

5 to 9.9 

10 to 14.9 

15 or more 



2 2 0.74 

122 -5 1.84 

23263111 19 6.99 

6 7 7 12 13 6 3 1 55 20.22 



2 3 7 13 15 18 9 3 1 71 26.10 

1 3 6 7 7 11 10 6 51 18.75 

1 2 2 5 5 15 10 4 44 16.18 

33883 25 9.19 

13 16 24 42 48 46 46 29 8 272 100.01 



160 TRAITS OF GIFTED CHILDREN 

length) and chest circumference, gives a large percentage 
of overweight boys (35 per cent) and of underweight girls 
(54 per cent) . Only 4 per cent of the boys are underweight 
and only 10 per cent of the girls are overweight, according 
to the Dreyer limits. When the limits of the Dreyer tables 
are extended to 10 per cent above and 10 per cent below his 
norm, 0.3 per cent of the boys and 25.4 per cent of the girls 
are underweight; 22.4 per cent of the boys and 2.6 per cent 
of the girls are overweight. 

The Dreyer method should be elaborated to take into 
consideration age differences, sex differences, and differ- 
ences of chest formation. Such an investigation is being 
carried out by the writer. 

In order to formulate standards for evaluating the nutri- 
tional status of children, Von Pirquet 1 used as his index the 
pelidisi, which are based on the relationship between weight 
and sitting height. The formula is : 



V IQxweight in grams 
Sitting height in centimeters ~~ 1UU per cent 

Table 42 gives the pelidisi for the boys and girls included 
in this study. This table shows that the index for the boys 
taken as a group is 94 per cent and for the girls 95 per cent, 
with a range of 86 to 105 per cent for the boys and of 86 to 
107 per cent for the girls. According to the Von Pirquet 
standards, the majority of these children range between 89 
and 99 per cent with a standard deviation of more than 3 
per cent, which would indicate that the normal rating should 
be a zone rather than a definite percentage. 

For the 14 boys and 30 girls with pelidisi of 99 and above, 
all are above weight for height and age by the Baldwin- 
Wood tables. The average amount overweight for the boys 
is 34 per cent, for the girls 23.2 per cent. For the 5 boys and 
6 girls below 89 in pelidisi, all are below weight for height 
and age. The average amount underweight for the boys is 
11.5 per cent and for the girls 17.3 per cent. 

Pirquet, C. F., von: Sitzhohe tmd Korpergewicht. Ztsch. f. KinderlieiL, 
1916, XIV, 211-228. 



ANTHROPOMETRIC MEASUREMENTS 



161 







|1 


---.-. aa a a,aiie8ea'-- | 


pf. 






"^ 03 


1 ^ 15 


^ CO O 
5 +1 










<D S S 






* 


1 


s<0 ?l 






3 


i 
""""I 5 " """"" '" 


PI 






CO 


"" -.--.-- ,, 


52J 




i 


9 


- -.-j a --- , 


S 33 

XO CO O 
i 




o 


a 


.,.,, ,5; 


S w ^j 


Q 




o 


rH ^ J, ,0 N B rt H N 


pf, 


s 




. 


H ... , 


?, 


o 

t 




00 


-"---- IS 


8e t| 


2 fe 




- 


iHr-IH t-l-ieOi-(eOt-l 105 


P| 


^ o 

W M 
A % 




gf 


"-" - BB^aa IB 


?, 


W 
PH 




* 


r-t rH C4 -t 1 US 


a-| 


^efl 




r-t 


<S <0 T-l M 00 >< 00 COC4C4 | Jg 


P| 


O? 
tf 




3 


"-- [., - - IS 


s s s 


d< 




a 


NCOO!I 

T-I f"l 1 S 


K SS 8 


> 








^ S S 




CO 


3 




a w f| 







s 


rt ^COJ.^ COH JJg 


8^' 






o> 


- - ^.|^. M H |B 


8-5 






00 


| , , s 


g gfi 






fc- 


H H ^1 i-i r^ | O 


S 8 S3 






s 

<tt 


; SS8S88s88$fes * 8gg 8 S! 8 6$ 


li 






1 


i 


r^ 



162 TRAITS OF GIFTED CHILDREN 

DEVELOPMENT OF CARPAL BONES 

Roentgenograms 1 were available for 57 of the gifted 
children, 29 boys and 28 girls, from 9 to 12 years. The selec- 
tion of children was made at random. The mean total areas 
of the carpal bones for these children are given in Table 43, 
with the mean areas for a group of Iowa children of the 
same ages. 

TABLE 43 

MEAN AREA OF CARPAL BONES OF 57 GIFTED CHILDREN AND 
126 IOWA UNIVERSITY SCHOOL CHILDREN 

Gifted Group Iowa Group 

Area Area Area Area 

Age N right left Age N right left 

sq. mm. sq. mm. sq. mm. sq. mm. 

Boys 

9 6 644 639 9 16 668 680 

10 13 754 754 10 11 880 857 

11 6 921 916 11 20 947 949 

12 4 887 885 12 14 1166 1157 

Girls 

9 1 775 756 9 17 801 791 

10 17 819 821 10 10 798 787 

11 9 967 962 11 20 1058 1036 

12 1 718 704 12 18 1216 1216 

From an examination of Table 43 it will be seen that the 
areas for the gifted children are lower at each age, except at 
10 years for the girls, than for the Iowa group. This may be 
due to chance, since the numbers involved make a very 
inadequate sampling. It is necessary, therefore, to take 
account of the height of these children. Height measure- 
ments were available for 43 of the group, 22 boys and 21 
girls, and for all the Iowa children. The average heights for 
ages show that the California gifted boys are shorter than 
the Iowa boys at 10, 11, and 12 years, and the California 
gifted girls shorter than the Iowa girls at 11 years. 

While the difference between the two groups is partly 
explicable on the basis of differences in height, it cannot be 
wholly explained in this way. 

x The roentgenograms were made by Dr. Moore. The measurements of 
the photographs were made under the direction of the writer. 



ANTHROPOMETRIC MEASUREMENTS 163 

STAGES OF PHYSIOLOGICAL MATURATION 

Observations were made on this group of children by the 
two examining physicians for some of the principal changes 
in both sexes during adolescence. (See page 577.) Five stages 
were indicated on the physical measurement cards to be fol- 
lowed by the examiners. These are (1) lack of apparent sex 
development; (2) beginning sex development, indicated by 
a slight appearance of straight pubic hair; (3) enlargement 
of genitals for boys and breasts for girls; (4) noticeable pig- 
mentation and curl of pubic hair; (5) complete pubescence 
for boys and first menstruation for girls. 

The progressive stages of physiological development nec- 
essarily grade into each other without sharp and accurate 
lines of demarcation. They are not capable of exact quanti- 
tative measurement like height in inches or weight in 
pounds. The correlations between the stages of physiological 
maturity and height, weight, width of hips, and circumfer- 
ence of chest of the gifted children have been calculated 
and are given in Table 44. It will be seen that all the cor- 
relations are positive and that probably all of them are 
significant. The highest correlations between physiological 
maturity and height are +.529.077 for girls at 12 years, and 
+.498 .072 for boys at 13 years. The highest correlations 
between maturity and weight are +.498 .083 for girls at 
13 years, and +.435.078 for boys at 13 years. These cor- 
relations indicate that, for the given ages, the taller and 
heavier children are relatively more matured physiologi- 
cally than the shorter, lighter children. 

In order to determine to what extent the correlations 
were due to weight and to height, respectively, partial corre- 
lations were computed in order to determine the relation 
between physiological maturity and the physical measure- 
ments when the influence of weight and height are sepa- 
rately eliminated. These partial correlations are given in 
Table 45. 

The significant correlations here are the correlation 
+.411 .088 between physiological maturity and height, 
when the effect of weight is eliminated, for 12 year old 
girls; the correlation +.264.090 between maturity and 
height, when the effect of weight is eliminated, for 13 year 



164 



TRAITS OF GIFTED CHILDREN 



TABLE 44 

INTERCORRELATIONS BETWEEN FIVE STAGES OF PUBESCENCE (PHYSIO- 
LOGICAL MATURITY), STANDING HEIGHT, WEIGHT, CIRCUMFERENCE 
OF CHEST, AND WIDTH OF HIPS FOR 290 GIFTED CHILDREN 

Pearson r 

Physiological Standing Width 

Age No. Maturity Height Weight of Hips 



11 65 



Boys 

Standing Height 
Weight 
Width of Hips 



Standing Height 

12 57 Weight 

Width of Hips 

Standing Height 

13 49 Weight 

Width of Hips 

Girls 

Standing Height 

11 42 Weight 

Width of Hips 
Circumference of 
Chest 

Standing Height 

12 65 Weight 

Width of Hips 
Circumference of 
Chest 

Standing Height 

13 37 Weight 

Width of Hips 
Circumference of 
Chest 



+.304 .076 

+.324 .075 +.662 

+.278 .077 +.549 +.858 

+.373 .077 

+.305 .080 +.705 

+.339 .079 +.677 +.860 

+.498 .072 

+.435 .078 +.813 

+.448 .077 +.593 +.746 



+.360 
+.422 
+.271 



+.529 
+.371 
+.388 

+.295 

+.291 
+.498 
+.426 



.090 
.086 
.096 



+.708 

+.630 +.840 



+.470 .081 +.571 +.894 +.760 



.077 
.092 
.090 



+.624 

+.565 +.845 



:.097 +.395 +.873 +.813 



.101 
.083 
.091 



+.725 

+.678 +.866 



+.428 .090 +.651 +.950 +.790 



old boys; and the correlation -K437:.090 between maturity 
and weight, when the effect of height is eliminated, for 13 
year old girls. On the basis of these partial correlations it 
may be concluded that the taller girls at 12 years are more 
mature physiologically than the shorter girls, that the taller 
boys at 13 years are more mature physiologically than the 
shorter boys, and that the more mature girls at 13 are heavier 
than the less mature. 



ANTHROPOMETRIC MEASUREMENTS 



165 





A 








SW 




ft 






2 


^ o 










C3 






H 






^ 


c *a 




Q 


S 




1 


aft 

g-S 




JZJ 


S 




O 


OQ 









-P 


rt 


a 




> N 

e 

S 


Q 




I 


I 


FSiologic; 
laturity 




5 


O 


S 




X5** 1 




^ 


^ 


^fap 




H 




^ 


CN 


M 






3 


S 


1 






W 

PH 


fe 
O 










W 











H 






10 

Tf< 


BESCEK 


i 










5^ 


1 


P 
o 


W 




i 


sw 
*S 

Ss 
1| 




o 


to 




S 








O 




o 


SQ^ 




C/3 


w 


.y, 


"3 


a 




S 


i 




i 


I 










1 







5 

"53 08 










a 




S 




w 

* 


o 

OH 


"S 






S 

W 





3 






riAL CORRELATIONS 


G HEIGHT, WEIGHT 








ft 


" 


* 


1 



-f 



"5 



XO rH 
<M 



00 

ft 



CO 
ft 

S 



-M -M O 

^ 'Sjfi 'S^ 
"S "S S """^ 



I 






O 

s 



S5 



tao w tic w 

'ft ft "J^ ft 

CL> >M Q} rH 

W ffi W ffi 



^5 



fl n3 CJ Td fl 'd 

5^H 0$ J Ctf ?^H 

J>. *-> u> M J>- 

co^ C/JK- co p- 



OO OO 

SS8 



CO 
CJ 



CO t 
CN 



+++ +++ + 



o 

M 

O 



s 



^a ->' 



58 



+ 



+ 



J CO T*< rH CD "i 
> rH CN 'TtJ rH C 

- l'+ + + 



0) 

O 







*3.&g 
KM 2 



o^o 

0> rd 0) 

tfi CO Q 

-&g 

aw 



S 

S 



*i 



M ' 



0) ^D 
rH CO 



bn co o 

'8.&S 

tSw 



2 gS 



166 TRAITS OF GIFTED CHILDREN 

NATIONALITY 

Deviations of nationalities in height and weight. The 
boys and girls representing a preponderance of a particular 
nationality were selected, and the amount of deviation above 
or below " the norm in height for age, and in weight for 
height and age was calculated. The results are shown in 
Table 46 for the seven predominating nationalities. The 
Jewish and Irish boys are the heaviest for their height and 
age and the American and Jewish girls are the heaviest for 
their height and age. The American girls include, however, 
two unusual cases of 34 and 60 pounds overweight, respec- 
tively. 

TABLE 46 

DEVIATIONS FROM THE NORMS IN HEIGHT AND WEIGHT OF GIFTED 
CHILDREN OF SEVEN PREDOMINATING NATIONALITIES 

Deviation 

Height Weight Height Weight 

Nationality Inches Pounds Inches Pounds 

N Boys N Girls 

American 11 .22 +1.8 8 +.28 +11.8 

English 43 +.01 2.0 49 .01 + 1.8 

German 41 .11 +3.0 29 +.10 + 1.9 

Irish 14 .78 +3.6 7 +.60 + 0.1 

Jewish 31 .41 +7.2 23 .30 + 8.2 

Scandinavian 6 .90 0.7 12 +.23 0.0 

Scotch 12 -.08 +2.3 14 +.74 + 0.8 

Deviations of nationalities in cephalic index. The ce- 
phalic index is the proportion of the width of the head to 
the length. This index is frequently used by anthropologists 
for determining race classifications. The width varies from 
70 to 90 per cent of the length for normal individuals. Within 
these limits three general divisions are indicated; (1) the 
long head, or dolicocephalic, below 75 per cent; (2) the broad 
head, or brachycephalic, ranging from 80 to 87 per cent; 
(3) the intermediate form, or mesocephalic, ranging from 
75 to 80 per cent It will be noted (Table 47) that all of 
these children lie in the upper range of the mesocephalic 
and the lower range of the brachycephalic, excepting the 
Scandinavian boys, who approach the extreme brachy- 
cephalic type. 

For the American boys the indices range from 75 to 84 
per cent, for the American girls, from 76 to 86 per cent; for 



ANTHROPOMETRIC MEASUREMENTS 



167 



the English boys, from 71 to 88 per cent, for the English girls, 
from 74 to 87 per cent; for the Jewish boys, from 74 to 87 per 
cent, for the Jewish girls, from 73 to 89 per cent; for the 
Scandinavian boys, from 81 to 92 per cent, for the Scandi- 
navian girls, from 75 to 85 per cent; for the Scotch boys, 
from 74 to 83 per cent (with one exception at 90), for the 
Scotch girls, from 71 to 81 per cent (with one exception at 
85). This illustrates the fact that various types are found 
within a particular nationality. This is especially true of 
this group of children, where the classification is based on 
simply a preponderance of nationality. 

TABLE 47 

AVERAGE CEPHALIC INDICES OF GIFTED CHILDREN OF THE SEVEN 
PREDOMINATING NATIONALITIES 



Nationality 

American 

English 

German 

Irish 

Jewish 

Scandinavian 

Scotch 



N 

11 
43 
41 
15 
31 
6 
12 


Mean 

.80 
.79 
.80 
.80 
.81 
.86 
.81 


Boys 
S. D. 

.0256 
.0368 
.0285 
.0304 
.0337 
.0389 
.0391 


Cephalic 

P. E. of 
S. D. 

.0037 
.0027 
.0021 
.0038 
.0029 
.0076 
.0054 


Index 

N Mean 
8 .81 

49 .81 
29 .81 
7 .80 
23 .81 
12 .80 
13 .78 


Girls 
S. D. 

.0361 
.0335 
.0325 
.0334 
.0337 
.0292 
.0308 


P. E. of 
S. D. 

.0061 
.0023 
.0029 
.0060 
.0033 
.0040 
.0041 



MENTAL AND PHYSICAL STATUS 

Since the mental examinations of these children were 
given prior to the physical measurements, the mental ages 
were computed to correspond with the exact chronological 
age at the time of the physical measurement, on the assump- 
tion that the IQ's for the short intervals remained constant. 

In a study of the physical status of this group of gifted 
children, it is found that the group is, as a whole, physically 
superior to the various control groups used for comparison. 
In order to determine the exact amount of correspondence 
between mental superiority and physical superiority, it 
would be necessary to study a very large group of unselected 
children. This was not possible for this investigation. 
Tables 48 and 49 give the correlations that were calculated 
for the gifted group. 



168 



TRAITS OF GIFTED CHILDREN 



TABLE 48 

COEFFICIENTS OF CORRELATION BETWEEN MENTAL AGE AND HEIGHT 
AND WEIGHT OF 397 GIFTED CHILDREN 



Boys 

N 



10 

35 



11 
68 



12 

63 



13 

50 



Mental age-height, r =.162 .111 .154 .080 .096 .084 .329 .085 

N 35 67 61 50 

Mental age-weight, r =.437 .092 .043 .082 -.090 .086 .305 .087 



Girls 

N 



N 42 48 45 46 

Mental age-height, r =.148 .102 .340 .086 .076 .095 .043 .099 

N 42 48 45 46 

Mental age-weight, r =.164 .101 .144 .095 -.055 .100 -.195 .096 

TABLE 49 

TOTAL AND PARTIAL CORRELATION BETWEEN MENTAL AGE, CHRONO- 
LOGICAL AGE, AND SEVEN PHYSICAL TRAITS OF 
594 GIFTED CHILDREN 

Coefficients 

Pearson with age 

Boys Trait Pairs coefficients constant 

Mental age standing height .835 .012 .219 .036 

Mental age weight .697 .020 .051 .038 

Mental age width of chest .662 .022 .044 .038 

Mental age depth of chest .582 .025 .002 .038 

Mental age circumference of chest .658 .022 .052 .038 

Mental age breathing capacity .742 .017 .168 .037 

Mental age grip .699 .020 -.072 .038 

Mental age chronological age .941 .004 

Girls 

Mental age standing height 

Mental age weight 

Mental age width of chest 

Mental age depth of chest 

Mental age circumference of chest 

Mental age breathing capacity 

Mental age grip 

Mental age chronological age 

The coefficients (Table 48) for weight and mental age 
for 10 year old boys, height and mental age for 11 year 
old girls, standing height and mental age, and weight and 
mental age for 13 year old boys indicate a positive relation- 
ship. 



.876 


.009 


.211 


.038 


.745 


.018 


.035 


.040 


.773 


.016 


,081 


.040 


.656 


.023 


-.014 


.040 


.756 


.017 


.011 


.040 


.792 


.015 


,076 


.040 


.766 


.017 


-.090 


.040 


.946 


.004 







ANTHROPOMETRIC MEASUREMENTS 169 

The correlations (Table 49) between mental age and 
standing height for the mentally superior boys and girls, in- 
dependent of the effect of chronological age, are found to be 
+.219:.036, and +.211dz.038, both of them probably signifi- 
cant correlations. 

When the entire group is taken as a whole, and chrono- 
logical age is partialled out, it is found that there is a slight 
positive correlation between mental age and standing height 
for boys and girls (Table 49) , but none between mental age 
and the other physical measurements. 

That a significant correlation between mental age and 
standing height is obtained for this selected group in which 
the IQ's range from 140 upward, indicates that the correla- 
tion would be much higher with a large unselected group in 
which the range of IQ's is wider. It is not possible to predict, 
from the data presented here, whether the amount of corre- 
spondence between superior physical measurements, other 
than standing height and weight, and superior mentality 
would be significant. 

SUMMARY AND CONCLUSIONS 

1. The gifted California children as a group are above the 
best standards for American born children in physical 
growth status for average standing height and weight. They 
also excel in average standing height and weight other 
groups of California children studied by Barnes, Boas, and 
Faber. 

2. The gifted children deviate in a positive direction 
from the Baldwin weight-, height-, age-, breathing standards 
for American-born children, but 62 to 73 per cent deviate 
not more than 10 per cent above or below these norms. 

3. A large proportion have broad shoulders and hips, 
strong muscles, and well-developed lungs. 

4. In the 37 physical traits, the boys surpass the girls in 
the averages of all traits up to 12 years of age, except width 
of hips, length of legs, and weight, in which the girls begin to 
show superiority prior to this age. After 12 years the girls 
excel the boys in standing height, sitting height, stem length 
to sternal notch, chest measurements, hip measurements, 
and weight. 



170 TRAITS OF GIFTED CHILDREN 

5. The physical traits of the girls are more variable than 
those of the boys, with the coefficient of variability increas- 
ing slightly with age for both sexes. For boys and girls the 
least variable measurements are for head, standing height, 
sitting height, stem length, width of shoulders, and width of 
chest. The most variable are 4hose of circumference of 
wrist, chest, and arms, weight, and the psychophysical 
functions of breathing capacity and strength. 

6. These children exc^l the children of a control group 
in Oak Park, Illinois, in four selected physical traits, arm 
span, width of shoulders, width of hips, and grip. 

7. Various types of cephalic indices are found within 
particular nationality groups represented by these children, 
but the majority of the children are of the mesocephalic 
type. 

8. The coefficients of correlation between all of the physi- 
cal traits are positive and high for each age and for each 
sex, ranging from .322 to .851. The coefficients are higher for 
boys than for girls. 

9. When chronological age is made constant by means 
of partial correlations, the lowest coefficient for boys is 
.483, between breathing capacity and depth of chest, and the 
highest is .703, between standing height and weight. For 
girls the lowest coefficient is .466, between breathing capacity 
and depth of chest, and the highest is .824, between breathing 
capacity and circumference of chest. 

10. There is positive correlation between the standing 
height of fathers and sons, and of mothers and daughters. 

11. This group of children, measured by the Dreyer 
method of predicting weight, gives a large percentage of 
overweight boys and underweight girls according to the 
Baldwin-Wood norms; measured by the Von Pirquet nu- 
tritional rating, the majority of the children are grouped 
between 93 and 96. 

12. The mean areas of the carpal bones of 57 of the boys 
and girls taken at random are found, by means of roentgeno- 
grams, to be slightly below the averages of Iowa children of 
the same ages. 

13. All the coefficients of correlation between physical 
traits and stages of physiological maturity are significant for 
boys and girls of the ages 11, 12, and 13. The intercorrela- 



ANTHROPOMETRIC MEASUREMENTS 171 

tions between the various physical traits are all high and 
positive. This shows that taller, heavier, larger boys and 
girls mature earlier than smaller children of the same 
chronological age. 

14. The coefficients of correlation between mental age 
and standing height and weight for ages 10 to 13 years vary 
considerably for boys and girls. Although low, they are 
positive for weight for 10 year old boys, for height for 11 
year old girls, and for height and weight for 13 year old 
boys. 

15. When age is made constant for the entire group of 
children from 2 years to 15 years of age by means of partial 
correlation, a small but probably significant positive correla- 
tion is found between mental age and height for boys and 
girls, but no correlation is found between mental age and 
other physical measurements. 

16. The results of this investigation show that the gifted 
group is, as a whole, physically superior to the various 
groups used for comparison. 



CHAPTER VIII 



HEALTH AND PHYSICAL HISTORY 1 

This chapter will summarize data on health history for 
the main experimental group. The data came from three 
sources: (1) the Home Blank, (2) the School Blank, and (3) 
the history records of the physicians who gave the medical 
examinations. The data from all three sources relate to the 
main group of 643 children, but the subjects reported by the 
home, the school, and the medical examiners are not always 
identical. In each case the data were obtained for 90 per 
cent or more of the group, but the subjects for whom data 
were not available were not always the same. For practical 
purposes, however, the three kinds of data may be treated as 
three lines of evidence on the health conditions of a single 
group. 

The items of information called for in the three blanks 
were not always the same, but (1) and (3) covered in general 
much the same ground. When a given question was included 
in both Home and School Blanks it was ordinarily worded 
identically in the two cases. The three schedules are as given 
below : 

HOME BLANK (Part I) 

I. PHYSICAL DATA* 

Bid you keep a "baby book," or diary* of the child's early development? 

1. Length of pregnancy (months child was carried). Child's weight at birth. 

2. Exceptional conditions of birth (prolonged or severe labor, use of instruments, 

e t c .) 

3. Mother's health during this pregnancy 

4. If child was breast-fed, from what age to what age? 

If child was bottle-fed, from what age to what age? ~ 

5. Describe child's health during the first year 

6. First teeth appeared at about what age? First permanent teeth? 

7. Age of learning to walk alone (several steps) To talk (short sentences) 

8. (If girl) has menstruation begun? -At what age? 

(If boy) Has voice changed? At what age? 

9. Has child had adenoids?. When?. If removed, state when 

10. Tonsil trouble? .When? Jf tonsils removed, state when 

II. Persistent mouth breathing (at present) : none, slight, marked, extreme. 

(Underline) 

12. Has child had colds very frequently, frequently, occasionally, only rarely? 
(Underline) 

a Written with the assistance of Florence L. Goodenough, Dr. Albert H. 
Moore, and Dr. Edith Bronson. 

173 



174 



TRAITS OF GIFTED CHILDREN 



HOME BLANK Continued 

13. Have eyes been tested? Nature of defect, if any Wears glasses? 

14. Has child suffered from headaches? At what age? How often? 

15. Is hearing excellent, good, fair, poor, very poor? (Underline) 

16. Is nutrition apparently excellent, good, fair, poor? (Underline) 

17. Has child had serious digestive trouhle? At what age? How serious? 

18. Symptoms of general weakness, if any - 

19. Is child especially nervous? How shown? , _ 

20. Chronic stuttering? From what age to what age? How severe? 

21. St. Vitus* Dance? At what age? How severe? 

22. Habitual muscular twitching? At what age? - 

Describe ..... , 

23. Has child had marked fears? At what age Fears of what? 

24. Has child had night terrors? At what age? How often? 

25. Has child shown a marked tendency to worry? Over what? 

26. Usual hour of going to sleep? Of waking? Is sleep sound? 

Average length of time required to go to sleep? 

27. What accidents, if any, has child had? 

After-effects, if any ~ ~ * 

28. What surgical operations? Was recovery normal? 

29. Illnesses : Write yes or no ; give age, severity, and after-effects. 



Diseases or illnesses Yes or no Age 



Severity 



Any lasting after-effects 



Measles 










Mumps 










Whooping-cough 










Chicken-pox 










Scarlet-fever 










Diphtheria 










Other Diseases 































SCHOOL BIANHL (Part I) 
I. HEALTH 

1. Does child have adenoids? Adenoids removed? 

2. Are tonsils diseased? , .Tonsils removed? 

3. Persistent tendency to mouth breathing (now) : none, slight, marked, extreme. 

(Underline) 

4. Has child colds frequently, occasionally, only rarely? (Underline) 

5. Is vision (without glasses) normal, somewhat defective, very poor? (Underline) 
Nature of defect, if any Does child wear glasses? 

6. Does child have headaches frequently, occasionally, not at all? (Underline) 

7. Is hearing normal, somewhat defective, very poor? (Underline) 

8. Is nutrition apparently good, fair, poor? (Underline) 

9. Symptoms of general weakness, if any 

10. Organic diseases (as of heart, kidneys, lungs, etc.) 



11. Is child especially nervous? Symptoms of nervousness, if any.. 



12. Does child stutter?..... How severely? , 

13. Has child had St. Vitus' Dance? Has child muscular twitchings?". 

14. Is child excessively timid? Prone to worry? , Cause of worry?.... 

15. Other indications of imperfect health , ."* 

16. Is above information based mainly on examination by a doctor?,' 
Nurse's examination? Your own observations? 

17. Does the school have a health examination record of this child?...! .... 



HEALTH AND PHYSICAL HISTORY 175 

MEDICAL BLANK (History Section) 
ADDRESS Tel Birth wt 

FAMILY Father alive wel1 iij tuberculosis lues chr. illness 

dead cause date 

TW^OT.. alive wel1 iu tuberculosis lues chr. illness 

Mother: dead cauge date 

Brothers: No.... alive well ill dead 

Sisters: No alive well ill dead 

No. mother's Pt. is Miscar- J No at mos.; mos. ; (Induced 

pregnancies.... no riages \ mos. ; mos. ; mos. ;|Spontan 

Insanity Epilepsy Asthma Other 

Feeblemindedness Other nervous Hayfever hereditary 

Remarks : 



PRENATAL 

BIRTH Full term Birth 

HISTORY Premature : b. at mos trauma.. 

Remarks: 



INFANCY Asphyxia Hemorrhage Scurvy Digestive 

Cyanosis Convulsions Rickets Malnutrition 

Breast Bottle feeding Fresh milt Vegetables 

fed mos< begun at mos ' Cond. milk began at 

Ace at which Sat alone - First steps 

Age at which cut flrst tooth First 

Remarks: 



LATER HISTORY 

(Give Dates) Specific Measles Scarlet Pneumonia Other diseases 

Infections; Pertussis Diphtheria Influenza Complications 

Sore throats Otitis Colds Rheumatism Toothache 

Tonsilitis Adenitis Bronchitis Chorea Abscess 

Loss of weight .Fever .Night cries Exposure 

Persistent cough, -Night sweats Growing pains to Tbc 

Digestive Nervous Urinary 
disturbance disturbances disturbances 

Tonsillectomy 19 Other Fractures 

Adenoidectomy 19 operations Accidents 

SBX DEVELOPMENT Pubic hair Menstruation 

Changes in voice Masturbation 

Remarks : - - - - - 



HABITS C1 from p. m. alone? Day nap? Out-of-doors 

bleeps to a m W indow open? hrs hrs. daily 

_ regular _,. bed? Brushes teeth times daily 

Bowels consti p ate d Wets clothes? Full bath times weekly 

fast? Milk, pints daily .Meat daily? 

between meals? Green vegetables daily? Fresh fruit daily? 

Tea Candy Appro*, no. 

Coffee calories daily 

Remarks: 



176 TRAITS OF GIFTED CHILDREN 

MEDICAL BLANK (History Section) 

HABITUS,, , , (strong ., . , J well-balanced irritable Habit 

Muscularly ^ weak Nervously ^ u^a^e -..easily fatigued.... spas 



r I weak ....^ ervousi y | unstable -..easily fatigued.... spasms... 

Active .Manageable Gets on with 

Quiet. Unruly. other children 

MENTALITY Age School grade Usual marks 

Remarks: 



(See pages 216 ff. for a statement regarding the examining physicians and re- 
garding the conditions under which the examinations were made.) 

Each of the above three sources of data has its advan- 
tages and disadvantages. Doubtless the medical history as 
recorded by the physicians is the most reliable, as it was ob- 
tained from the parent in person (usually the mother) at the 
time the child was brought for the medical examination. 
Information thus obtained is likely to be both fuller and 
more accurate than can be obtained by written replies. The 
information obtained from the Home Blank is usually from 
the same source as the history data reported by the physi- 
cians (the mother), but is less complete. In both cases the 
data are subject to the ordinary errors of report due to faults 
of memory and of observation. The cooperation of parents 
was such that it is believed the question of intentional falsi- 
fication of report need not be raised. In judging the value of 
the information derived from parents it is necessary to take 
account of the superior intelligence and education of the 
families represented in our group, and the fact that in about 
30 per cent of cases the parents had kept a written record in 
the form of a "baby book," 

The School Blank was in nearly all cases filled out by 
the child's regular class teacher. The teacher's information is, 
of course, less complete than that of the parent, since it cov- 
ers a briefer period; but for the same reason it is less likely 
to be vitiated by errors of memory. The teacher has also the 
advantage that she is likely to know more jthan the parent 
about the health conditions of the average child, which 
gives her a better standard for judgment. Moreover, she is 
able in many cases to refer to records of school health exam- 
inations made by school doctor or school nurse. Finally, 
the circumstance which makes the school reports of special 



HEALTH AND PHYSICAL HISTORY 



177 



interest is the fact that comparable reports were obtained 
from teachers for a control group, as well as for the gifted 
group. The sources of the teacher's information in the case 
of the gifted and the control groups were as follows : 

Main Source of Information Gifted Control 

School doctor 8.4% 9.6% 

Doctor and nurse 1.2% 2.2% 

Doctor and teacher 7.3% 9.0% 

Nurse 13.3% 7.2% 

Nurse and teacher 16.4% 15.8% 

Doctor, nurse, and teacher 4.6% 8.3% 
Teacher (sometimes includes observation of 

parents) 48.8% 47.9% 

School has health record for 75.4% 79.4% 

The control children attended the same schools as the 
gifted attended and usually, but not always, the same class. 
This means that in a majority of cases the School Blank for 
a control subject was filled out by a teacher who had filled 
one for a gifted subject. Effort was made to secure a control 
group which would represent as fairly as possible the entire 
school population of grades 2 to 8. The choice was made on 
an arbitrary and objective basis. In each case the teacher 
was directed to select that child in the class whose chrono- 
logical age was nearest to the age-grade norm according to 
the following standards: 

Grade 2 
3 
4 
5 
6 
7 
8 

The above standards correspond very closely to the actual 
grade medians for the school population in the cities cov- 
ered. 

Approximately 800 teachers, distributed fairly evenly in 
the schools which were canvassed after January 1, 1922, 
were asked to fill out the School Blanks for control subjects, 
one each. Approximately 600 did so with the necessary 
promptness, and the number of blanks filled out with suffi- 
cient completeness to make them usable was 527. These 
were distributed by age and sex as follows: 



7 years 9 months 


8 


10 




9 


11 




11 







12 


1 




13 


2 




14 


3 





178 TRAITS OF GIFTED CHILDREN 

Age Boys Girls 

8 45 36 

9 38 43 

10 38 42 

11 39 35 

12 36 35 

13 38 33 

14 33 36 

Total 267 260 

Grand total 527 

As other control groups were used for other purposes, 
this one will be referred to as Control Group A. 

LENGTH OF PREGNANCY 

Verbal report of mothers to physicians for 591 children 
of main experimental group gave the following results : 

Premature Full Term Overtime 

Of 317 boys 3.5% 94.9% 1.6% 

Of 274 girls 5.5% 94.2% 0.4% 

Sexes combined 4.4% 94.6% 1.0% 

The 26 cases of premature birth were distributed as 
follows : 

6^1 mos. 7 mos. 7% mos. 8 mos. 8% mos. 8% mos. Total 

Boys 1 7 2 1 11 

Girls 12147 15 

Length of pregnancy was reported in the Home Blank for 
565 children of the same group. Data from this source gave 
3.7 per cent of births as premature (not over 8 calendar 
months) and 1.4 per cent as overtime (10 calendar months). 
These figures agree closely with those based upon verbal re- 
ports to the physicians. 

MOTHER'S HEALTH DURING PREGNANCY 

Data are available only from the Home Blanks on this 
point. The following summary of responses is given for 
whatever it may be worth : 

N "Excellent" "Good" "Fair" "Poor" "Very Poor" 

Boys 320 30% 49% 13% 6% 2% 

Girls 254 35% 46% 11% 4% 4% 



HEALTH AND PHYSICAL HISTORY 179 

There were 44 conditions of ill health reported in the 
Home Blank, including 9 cases of "nervousness" and 7 cases 
of "nausea." No other condition was reported more than 
three times. 

WEIGHT AT BIRTH 

Birth weight was reported in the Home Blank for 569 
children, exclusive of those prematurely born. The distribu- 
tions are as follows, taking weight in each case to nearest 
half pound : 

N 4% 5 5% 6 6% 7 7% 8 8% 9 9% 10 

Boys 317 2 1 19 10 27 51 48 38 39 26 30 



10% 11 11% 12 12% 13 13% 14 14% 15 

8 11 2 4 1 

N 4% 5 5% 6 6% 7 7% 8 8% 9 9% 10 

Girls 252 1 3 3 10 18 34 33 46 32 28 12 16 

10% 11 11% 12 12% 13 13% 14 14% 15 

672 1 

Median Mean S. D. 

Boys 81bs., 3.10 oz. 8 Ibs., 6.26 oz. 1 lb., 5.96 oz. 

Girls 7 Ibs., 15.20 oz. 8 Ibs., 1.74 oz. lib., 6.22 oz. 

There were 64 boys and 57 girls reported to the physicians 
whose birth weights were the result of hospital weighings. 
These gave for the boys a mean of 8 Ibs., 5.2 oz. (S.D. 1 lb., 7 
oz.) , and for the girls a mean of 7 Ibs., 12 oz. (S.D. 1 lb., 2 oz.) . 
The means for the hospital weighings are only a very little 
lower than those based on the reports in the Home Blank, 
but the variability of the former is considerably lower. It is 
probable that the hospital records are the more reliable, but 
it should be noted that in both cases errors of memory may 
have entered. However, there are 140 boys and 68 girls for 
whom parents had kept written records (chiefly of home 
weighings) . These gave the following results : 

N Median Mean S, D. 

Boys 140 8 Ibs., 9.8 oz. 8 Ibs., 5.4 oz. 1 lb., 6.02 oz. 

Girls 68 8 Ibs., 3.4 oz. 8 Ibs., 3.4 oz. 1 lb., 6.18 oz. 

Both median and mean are higher for this group than for 
either of the other groups. For all the groups they are sig- 



180 TRAITS OF GIFTED CHILDREN 

nificantly higher than the norm, as will be seen from the 
following comparative data : 

Mean, boys Mean, girls 

Taber, 644 hospital cases 7 Ibs., 11 oz. 7 Ibs., 5 oz. 

"Holt 7 " 8.8 " 7 " 2.6 " 

'Pearson, 1000 consecutive 

hospital births 7 4.8 " 7 " 1.8 " 

4 Bowditch, Eastern United States 7 " 8.8 " 7 " 3.7 " 

'Brit. Anthropometric Com- 
mission, 1,000 cases 7 " 1.6 " 6 " 14.6 " 

'Skelheway and Gilbert, 500 
South Australian infants of 
British descent 7 " 15.3 " 7 " 9.2 " 

7 L. S. Hollingworth, N. Y. 

hospital cases 7 " 6.4 " 7 " 1.6 " 

The mean birthweights furnished by Dr. Faber are of 
special interest for our present purpose for the reason that 
they are based upon San Francisco infants. Although these 
means are higher than most other investigators have found, 
they are still far below the means for the gifted group. 

It may be of interest to compare the average of the four 
means for American children reported by Faber, Holt, Bow- 
ditch, and Hollingworth with the average of the means for 
the three separate gifted groups (total reported in Home 
Blank, cases reported in Home Blank for whom a "baby 
book" was kept, and cases of hospital weighings reported to 
physicians) : 

Mean, Boys Mean, Girls 

Average of Faber, Holt, Bowditch 

and Hollingworth 7 Ibs., 8.75 oz. 7 Ibs., 3.22 oz. 

Average of three gifted groups 8 " 5.6 " 8 " 0.4 " 
Excess of gifted over the norms 13.6 " 13.8 " 

Excess in per cent 11.3% 12.0% 

a Harold K. Faber: A Study of the Growth of Infants in San Francisco 
with a New Form of Weight Chart. Archives of Pediatrics, April, 1920. 

3 Holt: Care and Feeding of Children, Appleton & Co., 1912. 

Karl Pearson : Data for the Problem of Evolution in Man. Proc. of the 
Royal Society of London, Vol. 66, p. 23, 1900. 

IBowditch, H. P.: Eighth Annual Report, State Board of Health, Massa- 
chusetts, 1877. 

'Report of the British Association for the Advancement of Science, 1883, 
p. 253. 

'Robertson, T. B.: Comparison of the Weights at Birth of British In- 
fants. University of Calif. Press, 1915. 

T Hollingworth, L. S.: Comparative Variability of the Sexes at Birth. 
Am er. Journal of Sociology, Vol. 20, p. 335. 



HEALTH AND PHYSICAL HISTORY 181 

The superiority of the gifted is about three-fifths of the 
standard deviation of the gifted. It is conceivable that some 
of this superiority may be due to a tendency for birth weights 
to be exaggerated in memory reports, but if this were the 
explanation, the mean should be lower for children for 
whom written records were kept. Such was not the case. 
The evidence seems to justify the conclusion that our gifted 
children are above the norm with respect to weight at birth. 

ABNORMAL OR UNUSUAL CONDITIONS OF BIRTH 

The physicians obtained information on this point for 
591 children of the main group. In the Home Blank 536 of 
the same group were reported on by parents. The data from 
the two sources agreed very closely, but as the verbal reports 
made to the physicians are doubtless the more accurate, 
these only are presented. It will be noted (Table 49a) that 

TABLE 49a 
CONDITIONS OF BIRTH 

Of 317 Boys Of 274 Girls 

Instrumental delivery 60 (or 19%) 33 (or 12%) 

Induced labor 2 

Prolonged labor (48-72 hrs.) 2 

Foot presentation ' 1 

Breech presentation 4 5 

Caesarean section 3 

Vaginal caesarean 1 

Twilight sleep 1 2 

Left arm broken 1 

Cord about neck 1 

Congenital hip dislocation 1 

Badly bruised 23 16 

Hematoma 1 1 

Angioma 1 

Paralyzed (12 hrs.) 1 

Scar left by instruments 1 

Total conditions 96 65 

More than one condition listed 33 21 

Total cases unusual or abnormal 63 (or 19.9%) * 44 (or 16.0%) 

19 per cent of the births of males, and 12 per cent of the 
births of females, involved instrumental delivery. Such fig- 
ures would suggest that the common belief regarding the 



182 



TRAITS OF GIFTED CHILDREN 



influence of this factor in the causation of mental defect 
may not be well founded. 

INFANT FEEDING 

The data summarized are for 589 cases reported in the 
Home Blank. The results are given separately for the nation- 
ality groups: (1) American, including all American born 
mothers except those of Jewish ancestry; (2) foreign born, 
including Canadians but excluding all of Jewish ancestry; 
(3) mothers of Jewish or partly Jewish ancestry. This group 
includes a few mothers who have not reported their ancestry 
as Jewish but who are believed to be of Jewish origin. The 
most important facts are given in Table 50. 

TABLE 50 
INFANT FEEDING 



Amer. born, For. born, 

non-Jewish non-Jewish 

(N=428) 

1. Breast fed only, 

entire period 44.2% 



2. Bottle fed only, 

entire period 8.9% 

3. Partly breast, 

partly bottle 46.3% 

4. Report not clear 0.8% 

5. Breast only, 
eight months or 
longer 50.0% 

6. Breast only for 
less than eight 
months 



64.7% 

4.4% 

29.4% 

1,5% 

60.3% 



Jewish 
(N=93) 

50.5% 
7.5% 

40.1% 
1.1% 



Total 



47.5% 
8.2% 



All of 

160 IQ 

or above 

(N=88) 

57.8% 

4.4% 



43.5% 37.8% 
0.8% 0.0% 



58.1% 52.5% 55.6% 



40.4% 33.9% 33.3% 38.5% 



40.0% 



The proportion of breast feeding is probably high for the 
social classes represented in this group. It is highest in the 
foreign born non-Jewish, next highest in the Jewish, and 
lowest in the American born non-Jewish. The figures of the 
last column, for cases 160 IQ or over, show a slightly but 
perhaps not significantly higher percentage of breast feeding 
than is found for the total group. The low percentage of 
bottle feeding might be interpreted either as an indication 



HEALTH AND PHYSICAL HISTORY 



183 



that bottle feeding is not favorable to superior mental devel- 
opment, or that these parents have, because of their own in- 
telligence, recognized the importance of breast feeding. The 
latter is probably sufficient to account for all the difference 
found. 

Valuable comparative data on the proportion of infants 
who are breast fed are given for 20,504 cases by Woodbury. 1 
Table 51 shows the per cent of Woodbury's cases and of 
our gifted cases that were breast fed for various periods. In 
the preparation of this table infants which were "partially 
breast fed" are included but given only half weight. For this 
reason the figures for the gifted do not exactly tally with 
those in Table 50. Woodbury's data are confined to chil- 
dren who lived twelve months or longer. 



TABLE 51 




PERCENTAGE OF BREAST FEEDING AMONG UNSELECTED AND 


GIFTED CHILDREN 




Unselected 


Gifted 


(WoodCbury) 


(Stanford Group) 


Breast fed 1 month or longer 90.1% 


92.4% 





2 






81.6% 


89.2% 






3 






75.8% 


83.9% 






4 






69.1% 


77.5% 






5 






65.4% 


73.8% 






6 






61.6% 


68.0% 






7 






55.0% 


61.4% 






8 






51.0% 


57.4% 






9 






46.7% 


54.2% 






10 






41.4% 


47.6% 






11 






37.9% 


41.3% 






12 






35.6% 


34.7% 



The above figures show a considerably higher percentage 
of breast feeding for the gifted. It should be noted, more- 
over, that Woodbury's data were obtained in industrial 
cities 2 in which there was a large proportion of foreign born. 
In fact, 21.2 per cent of Woodbury's mothers are so classified. 
Since the percentage of breast feeding is higher in the case 

Robert M. Woodbury: The Relation between Breast and Artificial 
Feeding and Infant Mortality. Amer. J. of Hygiene, Vol. 2, November, 

'Johnstown, Pa.; Manchester, N. H.; Saginaw, Mich.; New Bedford, 
Mass.; Brockton, Mass.; Waterbury, Conn.; Akron, Ohio, and Baltimore, Md. 



184 



TRAITS OF GIFTED CHILDREN 



of foreign born than of native born mothers (both in Wood- 
bury's data and our own), the superiority of our gifted is 
even greater than the above figures would indicate. 

Table 52 gives additional comparative data. The fig- 
ures from Dietrich are based upon 1,000 consecutive cases 
seen in private practice in Los Angeles, including no hos- 
pital or welfare cases. Dietrich does not state whether the 
83 cases in which breast feeding was supplemented by the 
bottle were included in the totals for breast fed. Mitchell's 
figures are based upon nearly 3,000 cases in a children's hos- 
pital for the years 1900 to 1915. The cases were not consecu- 
tive and the method of selection is not given. The fact that 
this was a hospital group would make it of doubtful value 
for comparative purposes. 1 

TABLE 52 
ADDITIONAL COMPARATIVE DATA ON INFANT FEEDING 



Breast fed 1 month or longer 






2 


t( 








3 












4 












5 












6 












7 












8 












9 












10 












11 












12 









Dietrich 

84.9% 
78.2% 

70.8% 
61.8% 
55.6% 
51.5% 
44 .4 % 
39.2% 



Mitchell 

65.3% 
54.8% 
47.9% 
44.8% 
42.1% 
38.3% 
36.4% 
33.9% 
31.2% 
29.3% 
27.4% 



Gifted 

92.4% 
89.2% 
83.9% 
77.5% 
73.8% 
68.0% 
61.4% 
57.4% 
54.2% 
47.6% 
41.3% 
34.7% 



The superiority of the feeding conditions for the gifted 
group is even more marked in Table 52 than in Table 51. 
Perhaps the figures from Dietrich are the most valuable 
of all for our present purposes, since they are consecutive 
cases found in private practice among the middle classes 
of one of the cities covered in our survey. Only 39.2 per 
cent of his cases, as compared with 57.4 per cent of the 
gifted, were breast fed eight months or longer. 

*Henry Dietrich, M.D. : An Analysis of a Series of Case Records Relative 
to Certain Phases of Breast Feeding. J. of Amer. Med. Ass'n., July 22, 1922. 

Graeme Mitchell, M.D. : The Duration of the Nursing Period in Women 
of the United States. J. of Amer. Med. A$$'n., May 27, 1916. 



HEALTH AND PHYSICAL HISTORY 185 

EARLY HEALTH 

The physicians report the following abnormal conditions 
during infancy: 

Dr. Moore Dr. Bronson 

139 boys 142 girls 178 boys 132 girls 



Asphyxia 


5 


4 


7 


2 


Cyanosis 


6 


6 


5 


3 


Hemorrhage 





1 








Convulsions 


5 


1 


2 


3 


Scurvy 





1 


2 





Rickets* 








9 





Digestive* 


8 


8 


28 


23 


Malnutrition 


8 


8 


10 


10 



Total 32 29 63 41 

Two conditions reported for 39 children. 
Three " " " 8 

Four " " " 1 child. 

* A subjective difference between the reports of the two physicians probably 
exists here. In the positive findings recorded in the medical blank, Dr. Moore 
reports several cases as "suggestive of early rickets." (See report on deformities 
of head and chest, pp. 224 and 233.) These cases, however, are checked under 
rickets as normal and therefore are not included in the above table. 

Of a total of 591 cases, 107 (or 18.1 per cent) suffered 
from one or more of the above conditions. 

In the Home Blank parents rated "health during the first 
year" as shown in Table 53. 

TABLE 53 
HEALTH DURING FIRST YEAR (HOME BLANK) 

12 345 

N Excellent Good Fair Poor Very Poor Median 

Boys 317 25.6% 44.5% 12.3% 13.2% 4.4% 2.1* 
Girls 254 29.9% 49.2% 5.5% 13.4% 2.0% 1.9* 
Total 571 27.5% 46.6% 9.3% 13.3% 3.3% 2.0f 

* Equivalent of good. t Good. 

Slightly more than half of the special conditions of ill 
health during the first year reported by the mothers were 
digestive disorders. 

EARLY DEVELOPMENT 

Table 54 summarizes tlie testimony of mothers to phy- 
sicians at the time of the medical examination with re- 



186 



TRAITS OF GIFTED CHILDREN 



spect to age of sitting alone, teething, first steps, and first 
words ("at least three"). The data are subject to the usual 
vitiation from faulty memory and from lack of uniform 
meanings attached to such expressions as "sitting alone," 
"first steps," etc. The very early records are especially ques- 
tionable, and for this reason the medians are probably more 
significant than the means. 

Below Table 54 are given for comparison the means, 
medians, and S.D's based upon reports made by parents in 

TABLE 54 

DATA ON EARLY DEVELOPMENT 
(Total for both physicians) 





Sat alone 


Months 


Boys Girls 


3 


4 6 


4 


22 29 


5 


57 39 


6 


81 77 


7 


30 27 


8 


11 8 


9 


12 5 


10 


1 2 


11 


1 


12 


1 


13 





14 


1 


15 


1 


16 


1 


17 





18 





19 





20 





21 





22 





23 





24 





Above 24 





Unknown 


97 78 


Mean 


5.96 5.88 


Median 


5.83 5,81 


S. D. 


1.46 1.70 


Home 


(Not re- 


Blank 


ported) 


Mean 





Median 





S. D, 


_ 



First words 


First tooth 


First steps 


( u At least three") 


Boys 


Girls 


Boys 


Girls 


Boys 


Girls 


5 


13 














23 


14 








1 





34 


42 








2 


1 


54 


45 








2 


5 


65 


56 


1 


1 


5 


5 


34 


25 


1 


4 


14 


10 


28 


18 


15 


12 


20 


36 


17 


10 


13 


14 


43 


43 


7 


9 


36 


32 


37 


44 


9 


6 


68 


58 


76 


67 


4 


4 


44 


50 


25 


8 


2 


2 


49 


44 


12 


8 





2 


31 


20 


4 


3 



I 



34 

7.18 
6.89 
2.25 



7.83 
7.42 
2.43 



14 
7 

14 
1 
3 



12 

4 
6 
2 



1 



27 



2 
1 

16 



10 



5 

2 

11 



2 
2 

54 



2 


8 



7,02 13.10 12.87 

6.67 12.87 12.72 

2.42 2.63 2.46 

"Several steps" 

N=310 N=253 

14.16 14.08 

13.84 13.78 

2.97 2.83 



11.74 
11.60 
3.38 
"Short sentences" 



33 

11.01 
10.74 
2.37 



7,70 
7.23 
2.55 



17,82 17.05 

18.03 15.83 

5.21 4.29 



HEALTH AND PHYSICAL HISTORY 187 

the Home Blank. Attention is called to the fact that the data 
from medical blanks and Home Blanks are not comparable 
on age of walking and talking, as the terms were, unfor- 
tunately, differently defined in the two cases. 

Mead 1 reports the following data on age of walking and 
talking for normal and feebleminded children : 

Walking Talking 

N Mean A. D. N Mean A. D. 

Normal 50 13.88 mos. 1.56 mos. 50 15.32 mos. 3.0 mos. 

Feebleminded 144 25.08 " 9.6 " 92 38.52 " 16.8 " 

Comparison of means shows that our gifted children 
walked about one month earlier and talked about three and 
a half months earlier than Mead's normal children. Mead's 
data were based on the following definitions : 

Walking means "to take a step unassisted." 
Talking means "to use a word intelligently; i.e., to asso- 
ciate the idea with the object." 

The data for our gifted group show the girls slightly more 
precocious than the boys in sitting alone, teething, walking, 
and talking. In all cases, however, the difference is very 
small in comparison with the S.D. The coefficients of varia- 
tion, by the formula 

V= >X are as follows (for physicians 9 data) : 

Mean 

Sat alone First tooth First steps First words 

Boys 24.5 31.3 20.1 28.8 

Girls 28.9 34.5 19.1 21.5 

That is, the girls are more variable in two of the traits, and 
the boys in two. 

DISEASE HISTORY 

The most important data on this point are those of Table 
55, which is based upon the section of the medical exam- 
ination blank entitled "later history." 

1 C. D. Mead: The Relation of General Intelligence to Certain Mental 
and Physical Traits. Teachers' College, 1916, p. 117. 



188 



TRAITS OF GIFTED CHILDREN 



TABLE 55 

PARTIAL SUMMARY OF DISEASE HISTORY 
(Medical Examinations) 



Boys 
Dr. Moore Dr. Bronson 



Girls Totals 

Dr. Moore Dr. Bronson Both physicians 



Sore throat 


113(81.9%) 


100(56.2%) 


117(81.8%) 


88(66.7%) 


418(70.7%) 


Tonsilitis 


96(69.6%) 


83(46.6%) 


102(71.3%) 


79(59.8%) 


360(46.1%) 


Otitis 


35(25.4%) 


45(25.3%) 


30(21.0%) 


36(27.3%) 


146(24.7%) 


Adenitis 


27(19.6%) 


18(10.1%) 


12(8.4%) 


9(6.8%) 


66(11.1%) 


Colds 


126(91.3%) 


122(68.5%) 


128(89.5%) 


79(59.8%) 


455(76.9%) 


Bronchitis 


14(10.1%) 


40(22.5%) 


11(7.7%) 


22(16.7%) 


87(14.7%) 


Rheumatism 


1(0.7%) 


1(0.6%) 


4(2.8%) 


6(4.5%) 


12(2.0%) 


Chorea* 








1(0.7%) 





1(0.2%) 


Toothache 


15(10.9%) 


47(26.4%) 


13(9.1%) 


55(26.5%) 


110(18.6%) 


Abscess 


2(1.4%) 


24(13.5%) 





26(12.1%) 


52(8.8%) 



* One additional case of chorea was reported in the Home Blank. Both cases 
have entirely recovered. 

The total number of positive reports in the above table 
is 1,697. There were one or more positive reports for 550 of 
the 591 children of the main group who were given medical 
examinations. This is an average of somewhat more than 
three for each child. Reports of two or more illnesses were 
made as follows: 



Two 


Three 


Four 


Five 


Six 


19 


45 


38 


12 


4 


35 


36 


39 


16 


5 


20 


71 


24 


8 


2 


21 


24 


24 


22 


4 



Seven 

1 
1 
1 



Boys, Dr. Moore 
Boys, Dr. Bronson 
Girls, Dr. Moore 
Girls, Dr. Bronson 

Table 55 shows the large effect of subjective factors 
entering into medical statistics. Dr. Moore reports about a 
third more cases of sore throat, tonsilitis, adenitis, and 
colds than Dr. Bronson. That this difference is probably not 
due to climate or to hereditary tendencies in the population 
attracted to Los Angeles is suggested by the fact that Dr. 
Bronson reports bronchitis, another disease of the respira- 
tory tract, more than twice as often as Dr. Moore. Dr. Bron- 
son reports toothache nearly three times as frequently as 
Dr. Moore, and she reports 50 of the 52 cases of abscess. The 
disagreements are due in part to the greater tendency of Dr. 
Moore to ignore minor departures from perfect health, and 
in part to the fact that Dr. Bronson questioned the mothers 
somewhat more extensively than did Dr. Moore. 



HEALTH AND PHYSICAL HISTORY 189 

Table 56 summarizes the data of the physicians on cer- 
tain symptoms which are likely to be associated with tu- 
berculosis. 1 For all the symptoms in this group the figures of 
Dr. Moore are higher than those of Dr. Bronson. It is very 
probable that this is due to the fact that Los Angeles attracts 
so many families in which there is low resistance to tubercu- 
losis. The large figures for "growing pains" and "night 
cries" suggest that these symptoms are not very indicative of 
tuberculosis. The 35 children who were known to have been 
exposed to tuberculosis gave a smaller proportion of positive 
reports of growing pains than the total group. Night cries, 
however, were twice as frequent in the exposed group ; per- 
sistent cough, five times as frequent; fever, nearly three 
times; and night sweats, four times. 

TABLE 56 
HISTORY OF SYMPTOMS SOMETIMES ASSOCIATED WITH TUBERCULOSIS 

In total group : r>r - Moore Dr. Bronson Total 

Loss of weight 0.7% 0.3% 0.5% 

Persistent cough 1.4% 0.3% 0.8% 

Fever 1.1% 0.6% 0.8% 

Night sweats 3.9% 1.3% 2.5% 

Night cries 13.5% 1.6% 7.3% 

Growing pains 20.3% 4.8% 12.2% 
Per cent of total cases, 

minus duplications 34.9% 8.7% 21.2% 

Dr. Moore Dr. Bronson Total 

Cases exposed to tuberculosis: N=25 N=io N=35 

No symptom 64.0% 100.0% 74.3% 

Loss of weight 

Persistent cough 8.0% 5.7% 

Fever 4.0% 2.9% 

Night sweats 16.0% 11.4% 

Night cries 28.0% 20.0% 

Growing pains 16.0% 11.4% 
Per cent of exposed cases 

minus duplications 36.0% 25.7% 

Table 57 gives the per cents reported in the Home Blank 
as having had various infectious diseases. The second and 
fourth columns of this table give for each disease the per 
cent of attacks which were described by the parents as 
"severe" or "very severe/' or in words to that effect. 

'See p. 235 for results of lung examinations. 



190 



TRAITS OF GIFTED CHILDREN 



TABLE 57 
INFECTIOUS DISEASES REPORTED BY PARENTS (HOME BLANK) 



Measles 

Whooping cough 
Chicken-pox 
Mumps 
Influenza 
Scarlet fever 
Diphtheria 
Pneumonia 



S22 Gifted Boys 

Cases 

Have "severe" or 

had disease "very severe" 



88.8% 

67.8% 

61.8% 

37.6% 

19.3% 

9.1% 

5.9% 

4.3% 



19.6% 

24.7% 

4.5% 

9.1% 

37.1% 

24.1% 

26.3% 

48.0% 



262 Gifted Girls 
Cases 

Have "severe" or 

had disease "very severe" 



84.7% 

67.5% 

60.3% 

32.8% 

21.4% 

7.3% 

7.3% 

5.0% 



19.9% 
34.5% 
9.5% 
11.6% 
33.9% 
26.3% 
21.1% 
69.2% 



The above figures are almost identical for boys and girls. 
For both sexes the incidence of scarlet fever, diphtheria, and 
pneumonia seems high, but comparative data for the general 
population of the cities are not available. Roughly, one in 
twelve has had scarlet fever; one in fifteen, diphtheria; and 
one in twenty, pneumonia. About a quarter of the cases of 
scarlet fever and diphtheria and half of the cases of pneu- 
monia are described as having been severe or very severe. 
With an incidence so high, these diseases doubtless rob the 
world of many potential geniuses. At the same time, the 
frequency of severe cases among the superior children sug- 
gests that contagious diseases may not be as important a 
factor in the causation of mental defects as they are popularly 
believed to be. Other serious illnesses reported include 7 
cases of smallpox, 6 of typhoid, 3 of infantile paralysis, 7 
mastoid operations, and 11 appendix operations. 

"After effects" of contagious diseases reported by parents 
in the Home Blank include the following : 



Defect of hearing 
Eye defect 
Tonsil trouble 
Heart trouble 
Colds 

Weak joints 
Nervousness 
Bronchitis 



Cases Cases 

16 Loss in general tone 3 

15 Mastoid trouble 2 

9 Glandular trouble 2 

8 Chronic tonsilitis 2 

4 Indigestion 2 

3 Throat trouble 2 

3 Mouth breathing 2 

3 Anemia 2 



HEALTH AND PHYSICAL HISTORY 



191 



ACCIDENTS AND OPERATIONS 

Accidents were reported in the Home Blank as shown in 
Table 58. 

TABLE 58 
ACCIDENTS SUFFERED BY GIFTED CHILDREN (HOME BLANK) 



Accidents 



Boys 

No. After 

cases effects 

(cases) 



Girls 

No. After 

cases effects 

(cases) 



Total 

No. After 

cases effects 

(cases) 



Broken bones 


26 


2 


17 


1 


43 


3 


Motor accident including 














run-overs 


22 


4 


5 


1 


27 


5 


Guts, burns, and scalds 


14 


3 


13 





27 


3 


Sprains or dislocations 


6 


2 


1 





7 


2 


Miscellaneous 


37 


13 


30 


9 


67 


22 



Total 



105 



24 



11 171 



35 



After-Effects 








Impaired vision 


1 





1 


Heart murmur 


1 





1 


Minor deformities (fingers, 








toes, etc.) 


3 


1 


4 


Crooked arm 


1 





1 


Lameness 


1 


1 


2 


Spine affected 





1 


1 


Miscellaneous minor defects 


17 


8 


25 



Total 



24 



11 



35 



Surgical operations were reported by parents in the 
Home Blank as follows : 





Boys 


Girls 


Total 


Tonsillectomy 


157 


113 


270 


Adenoids 


161 


107 


268 


Circumcision 


57 


2 


59 


Appendectomy 


7 


4 


11 


Mastoid 


4 


3 


7 


Miscellaneous 


24 


15 


39 


Total 


410 


244 


654 



"Normal recovery" reported in all except 6 cases. 
"Slow recovery" reported in five cases of tonsillectomy. 
"Imperfect recovery" from Lorenz operation on hip (one 
case, child is still a cripple) . 



192 TRAITS OF GIFTED CHILDREN 

HEADACHES 

Information on the frequency of headaches was called 
for both in the Home Blank and the School Blank. For this 
condition, we therefore have data from two sources on the 
gifted, and from one source (School Blank) on a control 
group. "Frequent" headaches were reported as follows : 

Gifted Group . Control Group 

Home Blank School Blank School Blank 

Boys Girls Boys Girls Boys Girls 

2.0% 2.4% 2.3% 2.6% 4.2% 4.9% 

That is, home and school reports agree very closely on 
the gifted, but the school report shows almost twice as many 
control as gifted having "frequent" headaches. However, 
when the numbers for "frequent" and "occasional" were 
combined it was found that teachers report more gifted 
children than do parents as having headaches, but here also 
the gifted make a better showing than the control. From 
these figures it would seem that children are more likely to 
have at least occasional headaches at school than at home, 
and that this tendency is more marked with the control than 
with the gifted. Headaches may be largely psychological! 
Better physical care, less eyestrain because of more glasses, 
etc., may account in part for the differences between the 
control and the gifted, but it is doubtful whether it does so 
entirely. 

SYMPTOMS OF GENERAL WEAKNESS 

The School Blank called for symptoms of general weak- 
ness, if any. The question was answered for 527 of the gifted 
group and 594 of the control. Symptoms were reported as 
shown below. 

TABLE 59 
SYMPTOMS OF GENERAL WEAKNESS (SCHOOL BLANK) 

Gifted Group Control Group 

Boys Girls Total Boys Girls Total 
Number in 

group 321 273 594 269 258 527 

Pallor 44 88 5 13 

Nervousness 3 2 5 6 4 10 
Faulty nutrition 
(underweight, 

etc.) 7 8 15 11 4 15 



HEALTH AND PHYSICAL HISTORY 193 

TABLE 59 Concluded 

Gifted Group Control Group 

Boys Girls Total Boys Girls Total 

Posture poor 51 6426 

Anemia 224336 
Colds, sore 

throats, etc. 314033 

Spinal defects 12 30 55 

Fatigues easily 4 1 5 7 4 11 

Skin eruptions 101022 

Miscellaneous 11 5 16 8 3 11 

Total positive 41 26 67 47 35 82 
Per cent 

positive 12.8% 9.5% 11.3% 17.5% 13.6% 15.6% 

The above figures, it will be noted, are considerably more 
favorable for the gifted than for the control. 

URINARY DISTURBANCES 

Histories of urinary disturbances 1 were reported by the 
physicians as shown in Table 60. 

TABLE 60 
HISTORY OF URINARY DISTURBANCE 

Boys Girls 





Dr. 


Dr. 


Dr. 


Dr. 






Moore 


Bronson 


Moore 


Bronson 


Total 


Albumenuria 


1 


_ 





1 


2 


Cystitis 








2 


1 


3 


Pyelocystitis 


- 


- 


- 


1 


1 


Enuresis 


12 


3 


8 


2 


25 


Frequent urination 





_ 


1 


2 


3 


Painful urination 


_ 


1 


_ 





1 


Retention of urine 





- 


- 


1 


1 


Septic infarct 


_ 





- 


1 


1 


Stricture of urethra 


_ 


1 


- 





1 


Sugar in urine 





_ 


- 


1 


1 


Weak kidneys 


- 


_ 


- 


1 


1 


Indefinite 


1 


2 


- 





3 


Total reported 


14 


7 


11 


11 


43 


Per cent of 












total cases 


10.1% 


3.9% 


7.8% 


8.3% 


7.3% 



a See p. 244 for results of urine tests. 



194 



TRAITS OF GIFTED CHILDREN 



Reported positive 
Per cent positive 



DIGESTIVE DISTURBANCES 

Digestive troubles, past or present, were reported by the 
physicians as follows : 

Boys Girls Total 

Dr. Dr. Dr. Dr. 

Moore Bronson Moore Bronson 

22 40 24 28 114 

15.9% 22.5-% 16.9% 21.2% 19.3% 

The above figures agree fairly well with the data on di- 
gestive troubles reported in the Home Blank. According 
to the parents, 15.2 per cent of the boys and 12.5 per cent 
of the girls have had digestive disturbances. 

NUTRITION 

In the Home Blank nutrition was rated as "excellent/* 
"good," "fair," or "poor"; in the School Blank as "good," 
"fair," or "poor." Ratings from the two sources are, there- 
fore, not strictly comparable. In the following figures the 
most significant comparison is between the gifted and con- 
trol groups in the school reports. It will be noted that the 
school reports "poor" nutrition nearly three times as fre- 
quently in the control as in the gifted group. 





TABLE 61 








RATINGS ON NUTRITION 


N 


"Excellent" 


"Good" 


"Fair" 


"Poor" 


296 
242 
538 


~ ' ~ 


79.4% 
86.0% 
82.3% 


16.9% 
12.8% 
15.1% 


3.7% 
1.2% 
2.6% 


261 
251 
512 





71.6% 
80.5% 
76.0% 


19.6% 
13.9% 
16.8% 


8.8% 
5.6% 
7.2% 


315 
255 

570 


56.5 
63.5 
59.6 


36,5% 
29.8% 
33.5% 


6.4% 
6.7% 
6.5% 


0.6% 
0. 
0.4% 



Gifted Boys, School 
Gifted Girls, " 
All Gifted, 

Control Boys, " 
Control Girls, " 
All Control, 

Gifted Boys, Home 
Gifted Girls, " 
All Gifted, 



OBSTRUCTED BREATHING 

The Home Blank and the School Blank both called for 
information regarding removal of tonsils and adenoids. In 
the former, the question was answered for 550 of the gifted 



HEALTH AND PHYSICAL HISTORY 195 

group; in the latter, for 511 of the gifted and 493 of the con- 
trol. These reports are, of course, less accurate than those 
of the physicians but they are of interest because they were 
obtained for a control group in the cities covered by the 
survey. 

TABLE 62 
REMOVAL OF ADENOIDS AND TONSILS 

Gifted Gifted All Control Control All 

Boys Girls Gifted Boys Girls Control 

Adenoids Removed 

School Blank 44% 32% 39% 29% 18% 23% 
Home Blank 54% 42% 49% 

Tonsils Removed 

School Blank 48% 38% 44% 32% 18% 25% 
Home Blank 54% 44% 49% 

Three facts stand out in the above figures: 

(1) A far larger proportion of boys than of girls have 
had adenoid and tonsil operations. In this the school and the 
home agree. 

(2) Both in the case of adenoids and tonsils, the school 
reports about 60 per cent more removals for the gifted than 
for the control. 

(3) The home, as would be expected, reports more re- 
movals than the school, the latter overlooking about a quar- 
ter of all cases. 

Of 20,000 Denver children, 10 per cent were reported by 
teachers as having had adenoids or tonsils removed; of 
16,000 Salt Lake City children, 13 per cent. This is far lower 
even than the figures for our control group, a difference 
which is probably attributable to better medical attention 
given to school children in cities of California. 

MOUTH BREATHING 

Home and School Blanks called for a rating on mouth 
breathing as "none," "slight," "marked," or "extreme." 

The school reports 50 per cent more cases of mouth 
breathing in the control than in the gifted group, and three 
times as many cases which are "marked" or "extreme." 
This would be expected from the fact that the gifted have 
more often had adenoid and tonsil operations. 



196 TRAITS OF GIFTED CHILDREN 

TABLE 63 
HOME AND SCHOOL REPORTS OF MOUTH BREATHING 

N Slight Marked Extreme Total 

Gifted Boys, Home 285 28% 2% % 30 % 

Gifted Girls, " 239 17% 2% % 19 % 

All Gifted, " 524 23% 2% % 25 % 

Gifted Boys, School 266 21% 2% 1 % 24 % 

Gifted Girls, " 212 13% 3% % 16 % 

All Gifted, " 478 18% 2% 0.5% 20.5% 

Control Boys, " 228 35% 8% 1 % 44 % 

Control Girls, " 222 26% 5% % 31 % 

All Control, " 450 31% 1% 0.5% 38.5% 

There is a marked sex difference in favor of the girls. 
This is observed in both home and school reports, and for 
both gifted and control groups. A similar difference, in the 
same direction, was found with respect to frequency of 
adenoid and tonsil removal; that is, the girls have less often 
had tonsils or adenoids removed, and are less often mouth 
breathers. Somewhat more mouth breathers were reported 
by parents than by teachers. No consistent age tendencies 
were found. 

Since the gifted were classified into four grades by both 
home and school, it was possible to compute the correlation 
between the two ratings for extent of mouth breathing. This 
was done and found to be .56. 

Of 16,000 children in SaltLake City, 9 per cent were classi- 
fied by their teachers as mouth breathers. The school re- 
port for our gifted is 20.5 per cent, and for our control group 
38.5 per cent. The number of "marked" or "extreme" cases 
for our gifted group is only 2.5 per cent (school report), and 
for our control group 7.5 per cent. 

FREQUENCY OF COLDS 

Frequency with which the children suffered colds was 
reported in the Home and School Blanks as shown in Table 64. 

Again we find a marked sex difference in favor of the 
girls. In the gifted group more than twice as many boys as 
girls are said to have colds frequently or very frequently. 

The school reports a somewhat higher percentage of 
frequent or very frequent colds for the gifted than for the 



HEALTH AND PHYSICAL HISTORY 197 

TABLE 64 
FREQUENCY OF COLDS 

Freq. or 
N very freq. Occasionally Rarely 

Gifted Boys, Home 307 16% 47% 37% 

Gifted Girls, " 253 7% 44% 49% 

All Gifted, " 560 12% 46% 42% 

Gifted Boys, School 279 13% 30% 57% 

Gifted Girls, " 222 6% 19% 75% 

All Gifted, " 501 10% 25% 65% 

Control Boys, " 245 9% 35% 56% 

Control Girls, " 241 7% 34% 59% 

All Control, " 486 8% 35% 57% 

control, but this is offset by a higher percentage of gifted re- 
ported under the caption "rarely." 

Of 16,000 children in Salt Lake City, 20 per cent were re- 
ported by teachers as having colds as often as two or three 
times a month. 

HEARING 1 

Per cents rated in the Home and School Blanks as "some- 
what defective," "poor," or "very poor" in hearing are as 
follows : 

Gifted, both sexes, Home Blank 3.6% 

Gifted, both sexes, School Blank 2.3% 

Control, both sexes, School Blank 5.9% 

That is, parents report more cases of defective hearing 
among the gifted group than do teachers, and teachers re- 
port nearly three times as many cases for the control group 
as they report for the gifted. The difference is probably 
large enough to be significant, and may be related to the 
fact that more of the gifted group have had adenoids and 
tonsils removed. 

Data were worked out for the sexes separately, but no 
significant differences were found. 

A similarly worded question used by the writer in school 
surveys gave 4 per cent of 20,000 children with defective 
hearing in Denver, and 5 per cent of 16,000 in Salt Lake 
City. 

'See p. 224 for results of hearing tests. 



198 TRAITS OF GIFTED CHILDREN 

VISION" 

The School Blank called for a rating of vision (without 
glasses) as normal, somewhat defective, or very poor; also 
for information as to whether the child wore glasses. Com- 
bining ages, we have the following proportions of gifted and 
control groups rated either as "somewhat defective" or as 
"very poor" : 

Gifted Control 

Boys Girls Both Boys Girls Both 

Subnormal vision 19.9% 20.7% 20.3% 16.5% 15.4% 16.0% 
Wearing glasses 10.9% 10.3% 10.6% 4.0% 5.4% 4.7% 

About a quarter more cases of subnormal vision are reported 
for the gifted than for the control. This may be due to the 
fact that the gifted use their eyes more for reading, writing, 
and other near work. A more probable explanation is that 
with a given degree of defect the gifted, because of the 
greater intelligence of their parents, are more likely than 
other children to have the vision corrected by glasses; this 
would call the teacher's attention to the existence of a defect 
and cause it to be reported. 

NERVOUS DISTURBANCES' 

Under "later history" the physicians report 45.1 per cent 
of the boys and 32.3 per cent of the girls as having had a 
record of nervous symptoms, Dr. Bronson reporting about 
50 per cent more than Dr. Moore. One-third of all cases are 
accounted for by nail-biting. The remaining two-thirds are 
distributed widely among such symptoms as restless, "ner- 
vous," excitable, headaches, twitching, restless sleep, grind- 
ing teeth, sensitiveness, etc. 

Both Home and School Blank contain the question, Is 
child especially nervous? We thus have reports from both 
sources on the gifted, and from one source (School Blank) 
on a control group. The data are as follows: 

Gifted Control 

Boys Girls Total Boys Girls Total 

(N = 295) (N = 240) (N=535) <N^284) (N-243) (N = 527) 

School Report 16.3% 9.6% 13,3% 15.9% 16.4% 16.1% 

Home Report 24.7% 15.0% 20.4% 

^ee p. 226 for results of vision tests. 

2 See p. 243 for nervous conditions found in the medical examinations. 



HEALTH AND PHYSICAL HISTORY 199 

The parents report about 50 per cent more cases of ner- 
vousness than the teachers. Teachers report about the same 
number of gifted and control boys as nervous, but about 75 
per cent more control than gifted girls. 

In a survey of the Denver schools approximately 10 per 
cent of about 20,000 children enrolled were described by 
their teachers as showing such symptoms as muscular 
twitching, nervousness, excessive timidity, tendency to cry 
or to worry, stuttering, etc. The same method in a survey of 
the Salt Lake City schools gave 11.8 per cent of about 16,000 
for whom the question was answered. 1 Teachers in Philadel- 
phia reporting on 4,000 children classified 11.4 per cent of 
the boys and 9.6 per cent of the girls as nervous. 2 There is 
nothing in the above data to indicate that gifted children are 
more likely than others to show the ordinary symptoms of 
nervousness. 

In response to the question, How shown? 41 different 
symptoms of nervousness were mentioned. Of these, rest- 
lessness, excitability, irritability, and nail-biting account for 
more than half the cases. The other symptoms include cry- 
ing without cause, twitching, timidity, stuttering, worry, 
sensitiveness, trembling, restless sleep, etc. No significant 
differences were found between the gifted and control 
groups in the nature of the symptoms shown. 

Information on stuttering was asked for specifically in 
both Home and School Blank; also information regarding 
its severity. The results are as follows : 

Gifted Control 

Boys Girls Total Boys Girls Total 

School Blank 4.1% 0.8% 2.6% 6.0% 0.8% 3.4% 

Home Blank 2.9% 0.9% 2.0% 

Omitting the "slight" and "very slight" cases, we have the 
following : 

Gifted Contro 1 

Boys Girls Total Boys Girls Total 

School Blank 1.4% 0.7% 0.8% 0.4% 

Home Blank 1.1% 0.6% 

*The Denver and Salt Lake City data were collected by the writer by 
the use of questionnaire which was filled out by every classroom teacher 
in each city. The questionnaire asked the number of children in the class 
who showed symptoms of various kinds of defects. 

2 W. S. Cornell: Health and Medical Inspection of School Children, 1912, 
p. 595. 



200 TRAITS OF GIFTED CHILDREN 

That the majority of all cases reported are not very seri- 
ous is indicated by the fact that of the gifted group only 1.0 
per cent are reported as stutterers by both home and school. 

Of 20,000 children in Denver, 3 per cent were reported 
by their teachers as stutterers; of 16,000 children in Salt 
Lake City, 1.8 per cent. Conradi's census of 87,000 children 
in various cities of the United States gave 2.46 per cent with 
speech defects and 0.87 per cent as stutterers. There is no 
evidence in the above figures that stuttering is more common 
among gifted than among normal children. 

The teachers report no case of chorea for either the gifted 
or control group. The parents report 2 in the gifted group as 
having had an attack several years previously. Two cases in 
more than 500 represent about the normal frequency. 

Information regarding marked fears was asked for only 
in the Home Blank, hence no control data are available. Of 
the gifted boys, 10.3 per cent were reported as having 
"marked fears"; of the gifted girls, 13.0 per cent. Approxi- 
mately 80 per cent of the cases were under the age of eleven 
years. As causes of fear, darkness is mentioned 20 times; 
dogs and fire, 6 each; other animals than dogs, 5 times; 
nothing else more than twice. 

School reports were secured on excessive timidity for 
gifted and control groups. The results are as follows : 

Gifted Control 

Boys Girls Total Boys Girls Total 

Number reported on 284 228 512 242 230 472 

Per cent timid 4.9% 10.5% 7.4% 6.6% 7.4% 7.0% 

It will be noted that although there is little difference be- 
tween the gifted and control group, sexes combined, the 
gifted girls are reported "timid" twice as frequently as 
gifted boys. In the control group the difference is smaller, 
but in the same direction. 

The question in the Home Blank asking whether the 
child had had night terrors, and how often, brought positive 
reports for 9.4 per cent of the boys, and for 10.5 per cent of 
the girls. However, only 20 of the 51 who had had night ter- 
rors were subject to frequent attacks. This is out of a total 
of 515 reported on. The proportion having had frequent 
attacks is, therefore, 4 per cent. More than three-fourths of 



HEALTH AND PHYSICAL HISTORY 201 

all cases occurred before the age of eight years. More than 
half were described simply as "bad dreams" or "nightmares'* 
with screaming or crying on waking. 

Tendency to worry was reported as follows in the Home 
and School Blanks : 

Gifted Control 

Boys Girls Total Boys Girls Total 

School Blank 9.2% 12.0% 10.4% 7.8% 10.1% 9.0% 

Home Blank 8.8% 9.8% 9.3% 

No significant difference is found between the gifted and 
control groups, or between the home and school reports on 
the gifted. There were no marked age differences. School 
work was given as the source of worry in about half the 
cases in both gifted and control groups. 

HABITUS 

The following data on habitus are summarized from the 
medical blanks and are based in part on the observations of 
the examining physicians and in part on information re- 
ported to them by the parents. 

TABLE 65 
SUMMARY OF PHYSICIANS' REPORTS ON HABITUS 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

Muscularly strong 88.5% 95.8% 87.0% " 91.0% 

Muscularly weak 11.5% 1.4% 14.6% 9.1% 

Well-balanced 92.0% 97.2% 72.5% 71.2% 

Unstable 8.2% 2.8% 28.1% 28.8% 

Irritable 18.0% 7.1% 18.0% 12.9% 

Easily fatigued 10.8% 6.3% 12.4% 14.4% 

Active 84.1% 89.4% 80.3% 79.5% 

Quiet 15.8% 9.9% 18.0% 19.7% 

Manageable 97.1% 96.5% 93.8% 92.5% 

Unruly 2.9% 0.7% 6.2% 6.1% 

The highly subjective nature of the judgments on which 
the above figures are based is indicated by the large differ- 
ence in the reports of the two physicians on certain items. 
Dr. Bronson, for example, reports 28.8 per cent as "unstable," 
as compared with 2.8 per cent reported by Dr. Moore. She 
also reports nearly four times as many "unruly" as are re- 
ported by Dr. Moore. Dr. Moore reports three times as many 



202 TRAITS OF GIFTED CHILDREN 

boys as girls as "unstable," a sex difference which is not 
found in the data of Dr. Bronson. However, both physicians 
report more boys than girls as "irritable." Dr. Moore reported 
nearly twice as many children of the ages 10 to 13 as of the 
younger ages to be muscularly weak, unstable, irritable, eas- 
ily fatigued, and unruly, while in Dr. Bronson's data there 
were no significant age differences. 

EATING HABITS 

Eats fast. Of Dr. Moore's cases, 45.3 per cent of boys and 
22.5 per cent of girls were reported as "eating fast"; of Dr. 
Bronson's cases, 43.8 per cent of boys and 31.1 per cent of 
girls; of the entire group, 44,1 per cent of boys and 26.6 per 
cent of girls. In the reports of both physicians the boys show 
a small but probably significant increase as adolescence ap- 
proaches in the proportion of eating fast; the girls show a 
noticeable decrease. 

Eats between meals. Of Dr. Moore's cases, 62.5 per cent 
of the boys and 58.5 per cent of the girls are reported as eat- 
ing between meals; of Dr. Bronson's cases, 23.6 per cent of 
the boys and 16.7 per cent of the girls. However, in 36.8 per 
cent of Dr. Bronson's records the data on this point were not 
obtained. There were no marked age differences. 

Daily consumption of milk. The average daily amount 
of milk consumed by the boys is 1.12 pints; by the girls, 1.10 
pints. Of the boys 2.2 per cent are reported as drinking no 
milk; of the girls, 6.9 per cent. 

Green vegetables daily. This question is answered "yes" 
for 88.5 per cent of boys and 93.7 per cent of girls in Dr. 
Moore's group; and for 74.7 per cent of boys and 77.3 per 
cent of girls in Dr. Bronson's group. 

Eats meat daily. Dr. Moore reports "yes" for 60.4 per 
cent of boys and 59.2 per cent of girls; Dr. Bronson for 71.3 
per cent of the boys and 72.0 per cent of girls. It would be 
interesting to know whether the difference between the two 
groups is due to the more stimulating climate of the San 
Francisco Bay region where Dr. Bronson's cases live. 

Eats fruit daily. The record is "yes" for all but one of 
Dr. Moore's group, and for 84.3 per cent of Dr. Bronson's 
boys and 86.4 per cent of her girls. 



HEALTH AND PHYSICAL HISTORY 203 

Drinks tea. Of Dr. Moore's boys, 8.6 per cent are reported 
as drinking tea. In only two of these cases is the amount 
more than one cup a day; in three cases the tea is described 
as "very weak." Of the girls, 2.8 per cent are reported; in no 
case more than one cup a day. Of Dr. Bronson's boys, 9.5 
per cent drink tea at least occasionally. The amount is not 
specified. Thirteen, or 9.8 per cent, of Dr. Bronson's girls 
drink tea. One 7 year old boy is reported, but no others 
under 9. 

Drinks coffee. In Dr. Moore's group the record is "yes" 
for 8.6 per cent of boys and 9.2 per cent of girls; in Dr. Bron- 
son's group, for 15.2 of boys and 17.4 per cent of girls. No 
boy under 9 and no girl under 7 was reported as drinking 
coffee, and in no case was the amount more than one cup 
daily. 

Eats candy. The data on this point can be taken as sug- 
gestive only, as the answers were recorded in rather indefi- 
nite terms. Rough groupings were made as shown below. No 
marked age or sex differences were found. 

Dr. Moore Dr. Bronson 

1. "None," "very little," "rarely," 

"once a week" 29.6% 24.8% 

2. "Occasionally," "some," "moderate 

amount," "two or three times a week" 58.7% 31.0% 

3. "Daily," "often," "too much," "con- 

siderable," "a good deal" 5.3% 39.0% 

4. Not answered 6.4% 5.2% 

PERSONAL HYGIENE 

Data on constipation are reported by the physicians as 
follows : 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

"Regular" 94.3% 90.0% 87.0% 88.6% 

"Constipated" 5.0% 8.4% 8.4% 9.0% 

No record 0.7% 1.6% 4.6% 2.4% 

The following figures summarize the frequency of brush- 
ing teeth, as reported to physicians by parents. The various 
age groups are combined, as no significant age differences 
were evident. 



204 



TRAITS OF GIFTED CHILDREN 



Number of times daily 






Not at 








Indefinite 


Dr. Moore 


N 


all 


Once 


Twice 


Oftener 


answer 


Boys 


139 


1.4% 


56.8% 


31. 1% 


1.4% 


8.7% 


Girls 


142 


1.4% 


51.5% 


42.3% 


0.7% 


4.1% 


Total 


281 


1.4% 


54.1% 


37.0% 


1.1% 


6.4% 


Dr. Bronson 














Boys 


178 


2.8% 


50.6% 


39.3% 


3.9% 


3.3% 


Girls 


132 


0.8% 


31.8% 


59.8% 


6,1% 


1.5% 


Total 


310 


1.9% 


42.6% 


49.1% 


4.8% 


2.6% 



The mean number of baths per week, by age and sex, as 
reported to the physicians, was as follows : 



Age 

Boys 

Girls 



Mean number of baths per week by age 

2-45 6 7 8 9 10 11 12 13 
3.7 2.7 3.4 2.6 2.5 3.0 2.2 2.2 2.3 2.2 
4.5 5.0 3.4 2.8 2.8 2.4 2.4 2.6 2.1 2.3 

SLEEP HABITS 



14 
3.4 
2.3 



The Home Blank called for the child's usual hour of go- 
ing to sleep, time required to go to sleep, and soundness of 
sleep. Table 66 gives the median number of hours of sleep 

TABLE 66 
HOURS OF SLEEP OF GIFTED CHILDREN 



Hours 2-3-4 



Age 

8 9 



10 11 12 13 



12 



11 



2 


2 


4 


4 


6 


2 


4 





1 





9 


9 


6 


7 


12 


5 


2 


5 


4 


1 


1 


5 


7 


17 


16 


32 


21 


16 


9 


1 


2 


4 


10 


14 


26 


31 


30 


33 


16 


6 


_ 


_ 


1 


3 


10 


17 


37 


30 


23 


6 


1 


__ 


1 


_ 


4 


6 


12 


10 


9 


4 


_ 


_ 


1 





2 


1 


9 


10 


5 


2 


_ 








_ 








2 


2 


2 


1 



10 

9% 
9 

8% 

8 __ _ _- _ 

7% _____!____ 

Mean 11 :47 10 :29 11 :09 11 :14 10 :58 10 :38 10 :31 10 :27 10 :26 10:15 

47 55 43 34 41 37 44 40 43 41 
SJD. min. min. min. min. min. min, min. min. min. min. 

Mean for 1 

2,692con4 11:1410:4110:4210:13 9:5610:00 9:36 9:31 

trol cases J 



HEALTH AND PHYSICAL HISTORY 205 

by age for the gifted group, together with the age means 
found by Terman and Hocking for 2,692 unselected children. 1 

Table 66 indicates that these gifted children sleep more 
on the average than do unselected children. The differ- 
ence is small at six years, but increases to about three- 
quarters of an hour by the age of 12 years. Terman and 
Hocking found that feeble-minded children of the ages 6 to 
13 sleep considerably less than normal children, but that 
feeble-minded adults sleep more than normal adults. 

"Average time required to go to sleep" is 7 minutes for 
the group. Sleep is reported by the parents as "sound" for 
98.9 per cent of all cases. 

The physicians report that 79 per cent of boys and 70 
per cent of girls, or 75 per cent of all, sleep alone; and that 
96 per cent sleep with window open. A daily "nap" is re- 
ported for 7 girls and 8 boys, all but two of whom are under 
8 years. 

HOURS OUT OF DOORS DAILY 

As would be expected, the boys spend somewhat more 
time out of doors than do the girls, but there are no signifi- 
cant age differences. In .Dr. Moore's group, the boys average 
3.5 hours daily out of doors; the girls, 3.3 hours. Dr. Bron- 
son's figures show an average of 2.6 hours for boys and 1.9 
hours for girls. There are 3 of Dr. Moore's group who spend 
less than one hour out of doors, as compared with 24 of Dr. 
Bronson's group. Of Dr. Moore's group, 23 spend five or 
more hours out of doors; of Dr. Bronson's group only one. 
(Data for 587 cases). 

SEXUAL DEVELOPMENT OF GIFTED BOYS 

Condition of the pubic hair was recorded by Dr. Moore 
in the case of 115 boys of 9 years and older, for comparison 
with Crampton's data from 3,835 boys in the high schools 
of New York City. 4 

Crampton takes as his criterion of completed pubescence 
the "appearance of the kink or twist" (in pubic hair) "which 
is definitely characteristic." When this appears he considers 

1 The Sleep of School Children, etc. J. of Educational Psych., 1913, 138- 
147; 199-208; 269-282. 

^American Physical Educational Review, 1908. 



206 TRAITS OF GIFTED CHILDREN 

that the boy has reached puberty. During the period which 
extends from the beginning of the growth of pubic hair up 
to the time when the kink appears the boy is said to be pubes- 
cent. 

Crampton's figures are based upon examination of 3,835 
high school boys in New York City during the years 1901-06. 
Of these boys, 98 per cent were American born, but in about 
40 per cent of the cases both parents were born abroad. He 
calls attention to the differences in sex development between 
the various racial groups only in a general way, except for 
the German group, in which, he points out, puberty tends to 
occur at a later period than with children of American born 
parents. 

The data for Dr. Moore's group are as follows : 

Cases 
Age examined Hair present Percent Hair kinky Percent 

9 16 1 6.3% 

10 21 0% 

11 29 5 17.2% 

12 27 12 44.4% 

13 14 13 92.9% 3 21.4% 

14 55 100.0% 5 100.0% 
15-16 3 3 100.0% 3 100.0% 

As will be seen from the figures below, the children of 
foreign born parents were much more numerous in Cramp- 
ton's group. 

Birthplace of Gifted Group Crampton's 

Parents (115 cases) (3835 cases) 

Germany 2.3% 22.6% 

British Isles 3.7% 12.9% 

Russia 5.1% 6.2% 

Austria-Hungary 1.0% 6.2% 

Scandinavia 1.8% 1.6% 

Italy 0.4% 2.2% 

France 0.9% 1.0% 

Miscellaneous 6.0% 2.6% 

UNITED STATES 78.7% 44.8% 



99.9% 100.1% 

Table 67 shows the proportions of post-pubescents, 
pubescents, and pre-pubescents in the two groups at suc- 
cessive ages. 



HEALTH AND PHYSICAL HISTORY 207 

TABLE 67 
ADOLESCENT DEVELOPMENT, GIFTED AND NORMAL BOYS 

Pubescents 
Post-pubescents (hair present Pre-pubescents 

(hair kinky) but straight) (no hair) 

Age Crampton Gifted Crampton Gifted Crampton Gifted 

12 6.0% 0.0% 15.5% 44.4% 75.0% 55.6% 

13 24.5% 21.4% 27.0% 71.4% 48.0% 7.1% 

14 53.0% 100.0% 26.0% 0.0% 21.0% 0.0% 

15 77.5% 100.0% 15.0% 0.0% 7.0% 0.0% 

16 94.0% 0.0% 4.0% 0.0% 1.5% 0.0% 

17 99.0% 0.0% 0.0% 

The above figures would indicate that the gifted boy 
tends to mature somewhat earlier than the average, but the 
numbers in the gifted group are too small to be more than 
suggestive. 

Condition of pubic hair was not noted by Dr. Bronson, 
but change of voice was recorded by both physicians for all 
boys of 10 years or older. The figures are as follows for the 
total of 221 cases. 

Total Number Per cent changing 

Age cases changing or changed 

10 59 1 1.7% 

11 59 4 5.8% 

12 51 17 33.3% 

13 30 14 46.7% 

14 96 66.7% 
15-16 3 3 100.0% 

SEXUAL DEVELOPMENT OF GIRLS 

Presence or absence of pubic hair was recorded by both 
physicians for all girls of 10 years and over. Dr. Moore's 
cases were examined by a woman assistant. Condition of 
hair is noted in some instances as scanty or profuse, straight 
or kinky, but not in all; hence no differentiation is made in 
the totals given below. Reports of the two physicians agreed 
closely and have therefore been combined. 

Total Pubic hair Per cent of 

Age cases present total 

10 43 3 7.0% 

11 47 16 34.0% 

12 39 23 59.0% 

13 38 31 81.6% 

14 99 100.0% 
15-16 3 3 100.0% 



208 



TRAITS OF GIFTED CHILDREN 



M 



17 or 
older 



3.6 



00 
00 



o 
oo 



da 



last b 



^ ^ 

O O C- CO 

CNI ++ TH 



* C 

5 CO TH 



S 08 



A 

1 

03 

OS 

ir 

S 

3 



1 

a 
1 



2 
S 



CB ;2 O CO CM OO 2 ^ 

ii^^rH ^ C^ "^Irt 



o 

2 



oo 
" 



CM CM 



CO TH C^ CO 

>^5 r^I ^ (3* 



O CO C 

^ S 



00 t^ O 

e O QO 



O 
cC . 

oo Jz; 
co o 



S 





* 

TH 



CO CQ 



1 










OO CO 



oo 
oo 
oo 







CO 



CO 


CO 

H 











c 
o 


hlf 




c 








Ij- 


11 


c fi 

li 


2 S 

o w 


.si 


* 

a 2 





M 


o p3 

ft TH S 

5* K" 


Girls of 16 or 
Horace Mann '. 


3 

sl 

"5 ^ 

| o 
rf 


Girls above 11 
Baltimore Co, 


American wo 
Boston Hospi 


C0 

o 

1 

> 1 


^owditch repo 



HEALTH AND PHYSICAL HISTORY 209 

Age of first menstruation for those who have already 
matured is as follows, for the "present age" groups taken 
separately. It will be noted that age is taken to last birthday. 

Age of first menstruation (taken to last birthday) 
Present age N 10 11 12 13 14 15 

10-11 43 1 

11-12 47 - 1 

12-13 39 1 1 9 

13-14 38 - 7 11 10 

14-15 9 - 1 4 2 

15-16 3 1 - - - - 1 

Table 68a gives comparative data for age of first men- 
struation of normal American girls. The data for the first 
four groups are taken from Bird T. Baldwin's Physical 
Growth of Children from Birth to Maturity, University of 
Iowa Studies, 1921, p. 190. These groups are described as 
coming from the middle and upper social classes. Group 5 
is reported by H. P. Bowditch, Massachusetts Board of 
Health Report, 1877, and Group 6 by Burlage, American 
Journal of Physiology, April, 1923. 

Probably the data for the first four groups are the most 
accurate. The following figures (Table 68b) permit a com- 

TABLE 68b 

COMPARISON OF GIFTED GIRLS WITH NORMS IN AGE OF FIRST 
MENSTRUATION 

Per cent who had matured before various ages 
N 11 12 13 14 

Baldwin's 1 four groups 388 3 27 97 233 

0.8% 7% 25% 60.1% 
Gifted, 11 years old 

or older 136 

Gifted, 12 years old 

or older 89 2 11 

2.2% 12.4% 
Gifted, 13 years old 

or older 50 1 9 24 

2.0% 18.0% 48.0% 
Gifted, 14 years old 

or older 12 1 2 68 

8.3% 16.7% 50.0% 66.7% 

1 The figures presented for Baldwin's four groups in Table 68b have been de- 
rived from the data in Table 68a. It will be noted that in Table 68a the age of 
first menstruation is given to the last birthday, while Table 68b gives the number 
who have matured before a given birthday. Accordingly, the number who ma- 
tured before 13, for example, is the sum of those who are recorded in Table 68a 
as having matured at 10, 11, and 12. 



210 TRAITS OF GIFTED CHILDREN 

parison between this combined group of 388 girls and the 
girls of the gifted group with respect to the number of those 
who were eleven years old or older who had matured before 
11 or before 12; of the number who were thirteen years old 
or older who had matured before 11, before 12, or before 
13, etc. 

The number of gifted girls who have reached 13 or 14 is 
too small to give the above figures a very high degree of re- 
liability, but as far as they go they indicate a tendency to 
considerably earlier maturity for the gifted than for unse- 
lected girls. For example, of gifted girls 13 years old or 
older, about half matured before the age of 13, as compared 
with a quarter of unselected girls. 1 

Mammary development of girls was recorded by Dr. 
Moore but not by Dr. Bronson. A positive report includes 
all cases in which the pubertal development of the mam- 
mary glands had begun. Because of the difficulty of inter- 
preting the records no attempt has been made to differen- 
tiate as to extent of development. No strictly comparative 
data are available. 



Present age 


9 


10 


11 


12 


13 or over 


22 


25 


20 


18 


20 


1 


14 


16 


16 


20 


4.5% 


56% 


80% 


89% 


100% 



Cases 

Positive 

Per cent positive 

Seven histories of masturbation were reported to the 
physicians, including 4 boys and 3 girls. The data on this 
point are probably very incomplete. 

SUMMARY 

1. Data on health history for about 90 per cent of the 
main gifted group were obtained from the Home Blank, the 
School Blank, and the medical examiner's case history rec- 
ords. By means of the School Blank comparative data on 
many points were obtained from a representative control 
group of corresponding age attending the same schools 
which the gifted children attended. 

2. Of the gifted group 4.4 per cent were born prematurely, 
2.7 per cent as early as eight months. In only 7.8 per cent of 
cases was the mother's health during pregnancy rated as 
"poor" or "very poor." 

Compare with similar data for the gifted high school group, p. 578. 



HEALTH AND PHYSICAL HISTORY 211 

i 

3. The mean birth weight was approximately three- 
fourths of a pound above the norm according to accepted 
standards. This excess amounts to three-fifths of the stand- 
ard deviation of the birth weights of the gifted. 

4. Approximately 19 per cent of the male births and 12 
per cent of the female births involved instrumental delivery. 

5. Only 8.2 per cent of the gifted were bottle fed during 
the entire period, while 47.5 per cent were breast fed only 
and 43.5 were partly breast fed. The proportion of breast 
feeding was considerably higher than for the general popu- 
lation, and was appreciably higher for the cases above 160 
IQ than for the entire group. 

6. Health during the first year was rated by the mothers 
as "excellent" or "good" for 74 per cent of cases, and as "very 
poor" for only 3.3 per cent. 

7. Age of learning to walk averaged about one month 
less, and of learning to talk about three and a half months 
less, than mean ages for normal children. Dentition was 
perhaps slightly precocious. 

8. Summary of contagious disease history shows no im- 
portant deviations of this group from the normal child 
population, unless perhaps a rather high per cent have 
had scarlet fever (9.1 per cent) and diphtheria (5.9 per cent.) 

9. Nearly a third of the group have suffered one or more 
accidents; about 8 per cent of the group, bone fracture. The 
number of surgical operations averaged slightly more than 
one per child, more than half of which were for adenoids or 
tonsils. 

10. About half as many of the gifted as of the control 
group, according to school reports, suffer frequent head- 
aches. 

11. Symptoms of "general weakness" were reported by 
the school nearly 30 per cent less frequently for the gifted 
than for the control group. 

12. The school reports nutrition as "poor" for 2.6 per 
cent of the gifted as compared with 7.2 per cent of the con- 
trol. This is in harmony with the results of metabolism tests 
reported in Chapter IX. 

13. More than half of the gifted group had undergone 
tonsillectomy, as compared with about a quarter of the con- 
trol group. 



212 TRAITS OF GIFTED CHILDREN 

14. "Marked" or "extreme" mouth breathing is reported 
only one-third as frequently for the gifted as for the control 
group. 

15. In frequency of colds, no significant difference was 
found in the school reports on the two groups. 

16. Defective hearing is approximately two and a half 
times as frequent among the control as among the gifted, 
according to school reports. 

17. The school reports about a quarter more cases of de- 
fective vision for the gifted than for the control. 

18. Indications of "nervousness" are reported by the 
school for 13.3 per cent of gifted and for 16.1 per cent of 
control. Stuttering, including mild cases, is reported for 
2.6 per cent of the gifted and for 3.4 per cent of control. Only 
two cases gave a history of chorea. "Excessive timidity" 
and "tendency to worry" were reported with about equal 
frequency in the gifted and control groups. 

19. The data on habitus are difficult to evaluate, because 
of lack of suitable control data. 

20. Case histories obtained by the medical examiners in- 
dicate that for the gifted group the dietary regime is above 
the average for the general child population. 

21. Approximately 8 per cent of the gifted group suffer 
more or less from constipation. 

22. The gifted children show significant excess of daily 
hours of sleep as compared with the Terman and Hocking 
norms. The excess is slight with the younger children, but 
amounts to about fifty minutes by age 12. 

23. The gifted boys spend on an average about three 
hours out of doors daily, the girls about two and a half 
hours. 

24. Pubescence, as indicated by amount and kinkiness of 
the pubic hair, occurs on the average somewhat earlier 
among gifted than among unselected boys. Owing to the 
small number of gifted boys above twelve years, this con- 
clusion is only tentative. 

25. Of gifted girls 13 years old or older, 48 per cent had 
menstruated before 13, as compared with 25 per cent for un- 
selected girls. 



HEALTH AND PHYSICAL HISTORY 



213 



FIGURE 9 
GROWTH AND DEVELOPMENT OF GIFTED AND UNSELECTED CHILDREN 



Scale 

Units 

60 



50 
40 
30 
20 
10- 



CD 



(2) (3) 



(5) (6) (7) 



(8) 



Unshaded Gifted Children 
Shaded Unselected Children 

(1) Mean birthweight of boys, In quarter pounds. 

(2) " " " girls " 

(3) Per cent breast fed eight months or longer, both sexes. 

(4) Age of walking, in half months, boys. 

(5) " talking " 

(6) Per cent of 12-year boys having pubic hair. 

(7) Per cent of girls menstruating before 12 years. 

(8) Daily sleep, in half hours, both sexes. 



214 



TRAITS OF GIFTED CHILDREN 



FIGURE 10 
PHYSICAL DEFECTS IN GIFTED AND CONTROL GROUPS 



40" 



30-* 



ZQ-- 



BOYS 




(1) (2) (3) (4) (5) (6) (7) (8) (9) (1O) 



40'- 



30" 



20-" 



10- 



GIRLS 




(2) 



(3) (4) (5) (6) (-7) 

Unshaded Gifted Group 
Shaded Control Group 



(6) &) (10) 



(i) 

(2) 
(3) 
(4) 
(5) 
(6) 
(7) 
(8) 
(9) 
(10) 



Per cent having frequent headaches. 

with symptoms of general weakness. 

of mouth breathers. 

who have colds occasionally or often. 

with poor or very poor hearing. 

with vision somewhat defective or poor. 

nervous. 

with speech defects. 

who are exceptionally timid. 

who show tendency to worry. 



CHAPTER IX 



MEDICAL EXAMINATIONS 1 
SUBJECTS 

Medical examinations were given by the two physicians 
to 783 children of the various gifted groups. The purpose of 
this, as of other divisions of our study, was two-fold: (a) to 
secure data that would contribute to a better understanding 
of individual cases; and (b) to secure a basis for generaliza- 
tions with respect to the health conditions of gifted children 
in general. In the interest of the latter purpose the present 
summary is confined to the medical examinations of 591 
gifted subjects of the main experimental group. The number 
of families represented is 502, or 87 per cent, of the total of 
578 in the main group. A small part of the loss of 13 per 
cent was due to Christian Science beliefs of the parents, the 
greater part to such causes as change of residence, illness 
in the family, difficulties in the way of bringing the child for 
examination, etc. Table 69 gives the distributions of the 

TABLE 69 
SUBJECTS OF MAIN GKOUP GIVEN MEDICAL EXAMINATIONS 







Dr. Moore 


Dr. Bronson 


Age 


Boys 


Girls 


Total 


Boys 


Girls 


Total 


Total 


2-3-4 


2 


5 


7 


1 


1 


2 


9 


5 


3 


2 


5 


4 





4 


9 


6 


3 


5 


8 


6 


3 


9 


17 


7 


7 


12 


19 


3 


5 


8 


27 


8 


9 


13 


22 


14 


7 


21 


43 


9 


16 


22 


38 


28 


20 


48 


86 


10 


21 


25 


46 


38 


18 


56 


102 


11 


29 


20 


49 


40 


27 


67 


116 


12 


27 


18 


45 


24 


21 


45 


90 


13 


14 


13 


27 


16 


25 


41 


68 


14 


5 


5 


10 


4 


4 


8 


18 


15-16 


3 


2 


5 





1 


1 


6 



Total 



139 



142 



281 



178 



132 



310 



591 



Examinations by Dr. Albert H. Moore and Dr. Edith Bronson. Statistics 
of results prepared by Goodenough and Terman. 

215 



216 TRAITS OF GIFTED CHILDREN 

591 subjects by age, sex, and examining physician. Age in 
Table 69 is age at last birthday preceding the medical ex- 
amination, and averages about a year greater than age at 
the time the subjects were first located and tested. 

PREVIOUS EXPERIENCE OF THE EXAMINING PHYSICIANS 

The professional records of the examining physicians 
are as given below : 

Dr. Moore: B.S., University of Minnesota, 1895; M.D., 
University of Minnesota, 1897; post graduate work in New 
York City about twelve months; post graduate in pediatrics 
in New York and Baltimore; eighteen years in general prac- 
tice in New York City and New Rochelle, N. Y.; Captain, 
Medical Corps of the United States Army, in 1918; Senior 
residency in pediatrics at the University of California 
through 1919-20; attending physician in pediatrics through 
1920-21 in San Francisco state and county hospital; at pres- 
ent a member of the Hollywood Medical Group, practice 
limited to children; member of the Clinical Staff of the 
Children's Hospital of Los Angeles ; member of the Attend- 
ing Staff of the Hollywood Hospital. 

Dr. Bronson: M.D., Johns Hopkins University, 1913; In- 
terne, Children's Hospital, San Francisco, 1913-14; Resident 
Physician, New York City Children's Hospital, Randall's 
Island, 1914-15; Resident Physician, Hospital for Sick Chil- 
dren, Edinburgh, 1915-16; Resident Physician, Paddington 
Green Children's Hospital, London, 1916-17; Resident Phy- 
sician, Pendlebury Children's Hospital, Manchester, 1917-18; 
Temporary Physician, Out-Patient Dept., Children's Hos- 
pital, Great Ormond St., London, 1918-19; Associate Physi- 
cian, Children's Hospital, San Francisco, 1919-23; Assistant 
Professor, University of California Medical School, 1923. 

It is evident from the above records that both physicians 
were exceptionally well qualified by training and experience 
for making the proposed examinations. Both were recom- 
mended for the work by the pediatric department of the 
University of California and Stanford University Medical 
Schools. Both are highly trained specialists in children's 
diseases. Dr. Moore has had much more experience in gen- 



MEDICAL EXAMINATIONS 217 

eral practice, but Dr. Bronson has had a somewhat more 
varied and extended hospital experience. In a comparison 
of the data reported separately by the two physicians in this 
chapter it is well to bear in mind this difference in their ex- 
perience, as it probably accounts for some of the differences 
in number and kind of defects reported. It was most for- 
tunate for the investigation that two such competent examin- 
ers were available. Both took a genuine research interest in 
the outcome and worked with the most painstaking devo- 
tion. 

EXAMINATION SCHEDULE AND PROCEDURE 

The examination schedule was planned in conference 
between the examining physicians and an advisory commit- 
tee of the Stanford University medical faculty. 1 It was based 
in large part upon a schedule which had been used for some 
time by Dr. H. K. Faber, of the Stanford pediatrics depart- 
ment. It is intended to provide for an examination of 
approximately one hour duration, about equally divided be- 
tween history and examination proper. The section of the 
schedule pertaining to history has been given in the preced- 
ing chapter. The section pertaining to the examination 
follows : 

PHYSICAL EXAMINATION 

GENERAL Degree of Apparent p . r ft i A I skln .............. 

IMPRESSION prostration ................ nutrition ................ racies ................ ^oior ^ m m .............. 

_ . Moist ............ Turgor ............ Pigment ........... _ T 

SKIN Eruption ................ Dry ................ Sw * at .............. Oedema .......... Vasomotor ......... 

Shape ................ Fontanelle ................ Bosses ................ Scalp .............................. 

112X0 Hair .................. Sutures ................ 

Rt ................... Hearing .............................. M. T. Rt ................. Left ........................ 

EARS Left... ..... .'.....'.Ingersoll watch at ........ Mastoid .................. Discharge .............. 

p Rt ............... Lids .................. Corneae ................ Irides ................ Vision .............. 

Left ............ Conjunctivae....Pupils ........ Equal....Regular ........ Discharge ........ 

VT Shape ................ Discharge ................ ,., A 

NOSE Obstruction.... Excoriation ............. Blood .............................................................. 

., Lips ........ Tongue....Rt. tonsil....Left tonsil ................ Pharynx .......................... 

MOUTH Gums ...... Teeth ..... ^ Q lst ...... No 2nd ...... Soft p alate ........ Hard palate ........ 



Larynx ................ Sternomastoids ............................................ Stiffness 

NECK Thyroid .............. Glands : ant ......... post ......... rt ......... 1ft ..... Retraction .............. 

Deformity ................ Expansion .................... Resp .................................................. 

CHBST Beading .................... Intercostal angle ........ rate .................... rhythm ................. 

t.i ** Breath.Rales,... 

Lungs : Percussion ........ Palpation ........ Auscultation ........ y O ice....Rhonchi. 

Heart : size ...... Impulse....Rate sitting ............ Apex .................. Pulm Mnrmur 

Shape....Thrill .......... Rhythm....Sounds Tricusp ..... Aort.. U1 " it " 

l For personnel of this committee see p. 12. 



218 TRAITS OF GIFTED CHILDREN 

Distension....Patterns Tenderness Oedema Hernia 

ABDOMEN Recti Masses Spasm Fluid (umbilical) 

Liver from to cm. below costal margin Rt. kidney 

Spleen palpable to....cm. below costal margin Lft. kidney 

Stomach Small intestine Appendix Cecuin 

Colon Sigmoid Rectum Bladder Perineum 

Prepuce Testes Vagina Hernia Discharge 

GENITALS p e nis Scrotum Clitoris Hydrocele....Swelling 

. J Cervical Lumbar Deformity n'T?* 

BACK Spine ^ Dorsal S acral Rigidity D E spine 

Arm* 5Rt * 5 Rt Vasomotor...,Tenderness....Atrophy .... 

ILeft. . Hands I Left Trophic Swelling Hypertrophy 

EXTREMITIES j Rt j Rt Vasomotor..,.Tenderness....Atrophy .... 

Legs I Left Feet I Left Trophic Swelling Hypertrophy 

n* Epiphyses Deformity Swelling 

BONES Diaphyses Tenderness swelling 

Redness Tenderness 

JOINTS Swelling Fluctuation 

MUSCLES Atrophy Fibrillation Paralysis Nodules " Shortening 

TENDONS Hypertrophy....Tender points.-.Pseudoparalysis.-.Ganglion. * " iu5 *- 

Skin reflexes....Pyramidal Pupillary reflexes....Mydriasis....Chvostek 
NEUROLOGIC Ten< i on reflexes tract signs Clonus Gait Romberg 

Co-ordination Tremor of spread fingers Habit spasm 

Neuromuscular Tic Choreiform motions... .Speech Other 

MENTAL Excitement Irritability Delusions Hallucinations 

ENDOCHIN GLANUS Thyroid Other abnormalities 

First morning spec Albumen 

UBINE Office spec Albumen 

BLOOD Hb. (Talquist) 

B. P. Lying down Standing 

BASAL METABOLISM 

RADIOGRAPHS Rt. wrist Left wrist 

SPECIAL EXAMINATION (Half-page blank here) 
POSITIVE FINDINGS (Full-page blank here) 

CONDITIONS OF EXAMINATIONS, AND PROCEDURE 

Dr. Moore: "All examinations were made in my office (Holly- 
wood Medical Group). Appointments were made for 10 and 11 A.M., 
and for 2 and 3 P.M. As nearly as possible the time on each was an 
hour, including the history. A work sheet was used and copy made 
at night. Sixty per cent of the examinations were made in the after- 
noon. First the history was taken, child not being present; then the 
physical examination was made, following the -order of the examina- 
tion blank. The child was stripped to the waist. Genitals were regu- 
larly inspected by me in case of the boys and by a nurse in case of 
the younger girls. The older girls were not examined in this respect/' 

Dr. Bronson: "Examinations were made daily each afternoon of 
the week, and on three forenoons. The average time, including write 
up, was nearer an hour and a half than an hour. The history was 
taken in the order called for in the blank and statements were written 
down at once, with remarks. Frequently additional facts in history 
were added during the examination. At the point of 'Habits' (in 



MEDICAL EXAMINATIONS 219 

history) the child started to undress. Urine specimen in office was 
obtained while child was undressing behind screen. The examina- 
tion began with the child wrapped in a sheet, lying down. The back 
and extremities were examined .in standing posture, young child 
nude, older with posterior view nude, except a few males (adults) 
who were not stripped. The order was as follows: general impres- 
sion, skin, head, eyes, nose, neck, chest, abdomen, genitals, neuro- 
logic, mental, endocrine glands, urine, blood pressure, back, ex- 
tremities, bones, joints, muscles and tendons, ears, mouth, blood. The 
child dressed just before the examination of ears and mouth." 

Nearly all of Dr. Bronson's examinations were made 
in her private office in San Francisco, though a few were 
made in the psychological laboratory at Stanford Univer- 
sity. Dr. Moore examined the children who lived in the vicin- 
ity of Los Angeles, Dr. Bronson those of the San Francisco 
Bay region. One extended conference of the physicians and 
advisory committee was held before the examinations began, 
and another after each physician had examined about 50 
or 60 children. Unfortunately, owing to the distance by 
which the two centers are separated (500 miles) more fre- 
quent conferences were not possible. The data from the 
two sources are doubtless somewhat less comparable than 
they would have been had more frequent detailed compari- 
son of procedure been possible. 

GENERAL IMPRESSION 

Apparent nutrition. Dr. Moore recorded in terms of ac- 
tual weight, using Wood's scale. Dr. Bronson recorded ac- 
cording to personal impression, as she did not weigh the 
children. As the height-weight index has been treated in 
connection with the anthropometric measurements (Chapter 
VII) it did not seem worth while to present in any detail the 
records of the examining physicians on this point. 

Dr. Moore reports 40 per cent of his boys as "normal," 29 
per cent as overweight, and 31 per cent as underweight. Dr. 
Bronson reports 36 per cent of her boys as normal, 21 per 
cent as underweight, 12 per cent as "below par," sometimes 
adding such descriptive terms as "flabby," "nutrition poor," 
etc. She reports 20 per cent as overweight, 5 per cent as 
"undernourished," one with feminine distribution of fat, and 
10 with indefinite description, as "nutrition only fair," etc. 



220 TRAITS OF GIFTED CHILDREN 

Of the girls, Dr. Moore reports 31 per cent as normal, 34 
per cent as underweight, and 35 per cent as overweight. Dr. 
Bronson reports 31 per cent as normal, 16 per cent as under- 
weight, 6 per cent as "under par/' 6 per cent as "undernour- 
ished," and 38 per cent as overweight. 

According to the above figures, between a third and a 
quarter of these children are underweight or undernour- 
ished. The school reports, summarized in the preceding 
chapter, show 15.1 per cent with "fair" and 2.6 per cent with 
"poor" nutrition, as compared with 16.8 per cent "fair" and 
7.2 per cent "poor" for a control group. 

Fades. Dr. Moore reports 97.5 per cent of his cases as 
"normal," "bright," or "very bright" in appearance. Three 
cases are reported as "dull" in appearance, and 2 as having 
"adenoids f acies." (All of these are boys.) 

Dr. Bronson reports 95 per cent of her cases as "normal." 
She reports 9 cases of "adenoid f acies," 2 with "circles under 
eyes," and 1 "f acies of chronic intestinal indigestion." Two 
additional cases are underlined without description. 

Color of skin. Dr. Moore reports one 7 year old girl as 
"very pale," one 13 year old girl as having "flushed cheeks," 
and one 9 year old boy as "pale," all others as normal. 

Dr. Bronson reports 5 girls and 4 boys as "yellow" or 
"sallow," 2 girls as "florid," adding that one of these has 
"blue nails." She reports one girl and 7 boys as "pale," and 
underlines 2 other cases without description. All other cases 
are checked as "normal." 

Color of mucous membrane. Checked as normal by both 
physicians in all cases. 

METABOLISM TESTS 

Dr. Moore secured metabolism tests of 93 subjects (47 
boys and 46 girls) selected at random from the gifted group. 
The age distribution of these subjects was as follows: 

Age Boys Girls Total 

6101 
7011 
8257 
9325 

10 6 9 15 

11 10 13 23 



MEDICAL EXAMINATIONS 221 



Age 


Boys 


Girls 


Total 


12 


14 


9 


23 


13 


9 


4 


13 


14 


2 


3 


5 



Total 47 46 93 

The tests were made by Mr. Calvin van Schaak, M.S., 
chemist and bacteriologist. The procedure is described by 
Dr. Moore as follows : 

The subject was ordered to eat a light supper the evening pre- 
vious to the test and report to the laboratory as conveniently as pos- 
sible in the morning with no breakfast, medication, water or other 
intake of any kind. On arrival, careful inquiry was made regarding 
this point and the test was deferred in those instances in which 
there was acknowledgment that anything had been eaten. Inspec- 
tion was then made for tight or uncomfortable clothing or hair ar- 
rangement, the bladder was emptied, temperature and pulse were 
noted, and the subject was then isolated at bed rest for at least thirty 
minutes under as quiet conditions as were possible in a busy group 
office. In most instances better control was established by keeping 
out parents and companions throughout the entire procedure once 
the subject was turned over to the laboratory. 

Following this the subject was transferred to the metabolism 
room and put comfortably to rest; the metabolimeter was demon- 
strated on the operator and the procedure was explained in so far 
as it could be appreciated. In most subjects, confidence, relaxation, 
and enthusiasm, with a spice of competition, was obtained quite 
readily, and although there were a few who were somewhat hesitant 
when it came to the final steps, none was so apprehensive as to ac- 
count for a marked pathologic rating. Those who showed definite in- 
stability were privileged to hold the mouth piece and nose clamp to 
themselves during the dummy test. The subjects were then allowed 
to rest alone until at ease and with a pulse rate practically the same 
as when it was first taken, usually in about 15 to 20 minutes. The 
test was then run for five minutes, after which a rest was allowed of 
from three to five minutes, depending on the subject. Some of the 
smaller subjects became tired and experience showed it best not to 
run the test in such cases longer than five minutes at a time. If this 
second test showed too great variance from the first, a third was 
made, but this was only essential in two or three instances. A few 
showed rates sufficiently high to warrant suspicion that food had 
been taken and these were repeated at a later date with more satis- 
factory results. A few frank abnormals were repeated but with prac- 
tically the same readings on the check test. 

All inquiries as to age and taking of height and weight were de- 
ferred until the tests were completed. 

The Sanborn portable metabolimeter was used throughout, and 
careful attention was given to see that the soda-lime was fresh and 
dry and that leakage was avoided. The temperature of the water and 



222 TRAITS OF GIFTED CHILDREN 

apparatus was noted before and after test, allowances were made for 
variations of barometric pressure, and a reliable stop watch was used 
for timing. The only outstanding difficulty encountered was due to 
inability to adapt the usual form of mouthpiece accompanying this 
instrument to some of the ill-formed mouths, necessitating special 
modifications. 

The following standards were used for interpreting the rates : 

Ages Sex Normal Standard Accepted 

6 to 12 M&F Benedict-Talbot 

12 to 14 F Benedict-Hendry 

12 to 14 M Modified DuBois 

Careful consideration was given each individual case as to regu- 
larity of respiration, bodily movements, avoidance of test during 
premenses, unstability, etc. 

The mean rating of the boys was 108.2 (S.D. 13.87) ; that 
of the girls, 111.9 (S.D. 1412). Following are the numbers 
below 90, between 90 and 110, and above 110. 



Hyper 
(AhovellO) 


Normal 
(90-110) 


Hypo 
(Below 90) 




N 


% 


N 


% 


N 


% 


Boys 
Girls 


21 
29 


44.7% 
63.0% 


22 
13 


46.8% 
28.2% 


4 
4 


8.5% 
8.8% 



Total 50 53.8% 35 37.6% 8 8.6% 

Dr. Moore writes: "I believe these figures to be fairly 
correct. The tests were made by a careful man who was in- 
terested in the work. The results seem to indicate that nearly 
all of these subjects are either normal or hypers as far as 
metabolism is concerned. Normal means within 10 per cent, 
either way, of the norm for age, height, and weight." 

When the metabolism ratings of these 93 children were 
compared with their IQ's, no significant correlation was 
formed, although the average IQ was slightly higher for the 
hypo group than for either of the other groups. The figures 
are as follows: 

Hyper Normal Hypo 

Mean IQ of boys 152.0 150.3 157.0 

Mean IQ of girls 150.9 152.7 154.5 

SKIN 

Conditions of the skin, including vasomotor disturb- 
ances, are summarized in the following table: 



MEDICAL EXAMINATIONS 223 

TABLE 70 
CONDITIONS OF THE SKIN 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

Normal in N % N % N % N % 

all respects 120 86.3% 121 85.2% 74 41.6% 37 28.0% 

Eruption 5 3.6% 5 3.5% 22 12.4% 27 20.5% 

Moist 4 2.3% 3 2.1% 7 3.9% 11 8.3% 

Dry 32 18.0% 27 20.5% 

Turgor 2 1.4% 1 0.7% 11 6.2% 7 5.3% 

Sweat 2 1.4% 3 2.1% 30 16.9% 48 36.4% 

Pigment 1 0.7% 3 1.7% 3 2.3% 

Vasomotor 14 10.1% 15 10.6% 59 33.2% 73 55.3% 
More than 

one condition 6 4.3% 3 2.1% 44 24.7% 65 49.3% 

The amount of disagreement in the reports of the two 
physicians is very marked. Dr. Moore records 85.7 per cent 
of his cases a;s normal, Dr. Bronson 35.1 per cent. Whether 
slight acne is recorded as an eruption; whether slightly 
"clammy hands" is recorded as a vasomotor disturbance; 
whether skin that is somewhat moist or dry is recorded as 
abnormal, seems to be largely a matter of the personal 
equation. 1 

Questions addressed to the physicians brought the following notes re- 
garding standards. 
Dr. Moore: 

Eruption includes acne, eczema, and impetigo. 

Moist, only when apparent to the touch. 

Dry, only a scaly condition. 

Turgor, normal unless a case of malnutrition. 

Sweat, a tendency to perspire in droplets during examination. 

Vasomotor, extreme redness or cyanosis of extremities.'* 
Dr. Bronson: 

Eruption. Pityriasis alba, due to exposure to sun, etc., is probably 
not so frequent in Bay region as in southern California. Acne vul- 
garis is common in adolescents. One instance of psoriasis. 

Dry. Malnutrition, especially if due to a chronic illness, produces a 
dry, rough skin. 

Turgor. Refers to tone or feeling of the flesh. It indicates water con- 
tent; for example, in a diarrheal baby the tissue turgor is much 
diminished. It is not a term of much significance with older chil- 
dren. 

Sweat. Not infrequently, especially in adolescents, the sweat dripped 
in the axilla during examination. This is most frequent in the 
"nervous" type. 

Vasomotor. A good test for this is the "Tache cerebral." A line drawn 
with the nail across the chest produces a red mark which persists 
for a considerable time. Blushing is also a sign of vasomotor in- 
stability. 



224 TRAITS OF GIFTED CHILDREN 

HEAD 

Dr. Moore reports only 2 cases of abnormal shape of 
head; one 7 year old girl with "large" head, and one 11 year 
old boy with "long, narrow" head. He reports bosses for 
two 8 year old girls (identical twins) . No other cases of ab- 
normalities are reported by Dr. Moore for any of the items 
in this section. 

Dr. Bronson reports 8 girls and 17 boys as having "squar- 
ish" heads, and 4 girls and 5 boys with other abnormalities of 
shape, making a total of 11 per cent with abnormalities of 
shape. She reports "oily," "dry," or "coarse" hair for 24 
girls and 5 boys, or 9 per cent of her cases; depression of the 
f ontanelle for 14 girls and 26 boys, and other abnormalities 
of the fontanelle for 4 girls, making a total of 44, or 14 
per cent of all her cases; open sutures for one 11-year- 
old boy and bosses for 13 girls and 18 boys, or 10 per 
cent of her cases; unhealthy scalp conditions for 19 girls 
and 12 boys, or 10 per cent of her cases. The latter con- 
ditions are recorded chiefly as "dandruff," "very dry," "very 
dirty," etc. 

No significant age differences are noted in the reports of 
either physician. 

HEARING 

The watch test was used (ordinary Waltham watch) . Dis- 
tances were recorded, but because of varying acoustic condi- 
tions, noise in buildings, etc., these can be considered as only 
approximately correct. Each physician set a "normal" range 
for hearing, also ranges for rough groupings under the fol- 
lowing heads: decidedly defective, somewhat subnormal, 
normal, superacuity. 

No significant age differences were found. 

Dr. Moore reports one boy as having decidedly defective 
hearing in both ears, and 2 girls as having decidedly defec- 
tive hearing in one ear. 

Dr. Bronson reports 2 boys with decidedly defective hear- 
ing in both ears, and one girl with decidedly defective hear- 
ing in one ear. 

Dr. Moore reports 16 boys (11.5 per cent) and 26 girls 
(18.3 per cent) as having somewhat subnormal hearing in 



MEDICAL EXAMINATIONS 225 

one or both ears; Dr. Bronson, 11 boys (6.2 per cent) and 5 
girls (3.8 per cent) . 

Dr. Moore reports 122 boys (87.8 per cent) and 114 girls 
(80.3 per cent) as having normal or above average acuity of 
hearing in both ears; Dr. Bronson, 165 boys (92.7 per cent) 
and 126 girls (95.4 per cent) . 

The school reports only 2.3 per cent of gifted with hear- 
ing "somewhat defective," "poor," or "very poor," as com- 
pared with 5.9 per cent of the control group. 

EAR CONDITIONS 

Examination of tympanic membrane was made by Dr. 
Moore in all cases except 7 boys and 8 girls. All of these had 
excellent hearing, hence percentages are based upon total 
groups. Membrane is reported normal for 133 or 95.7 per 
cent of the boys, and for 134 or 94.4 per cent of the girls. 
Abnormalities were recorded as follows : 

Both drums dull, reflexes obliterated 1 

Drums or canals obscured by wax 4 

One or both drums retracted 5 

" " " " thickened 2 

" " " dull 2 

Dr. Moore reports one boy and one girl with scar left by 
mastoid operation, and one 10 year old boy as having "prob- 
able mastoid trouble" at present. He reports no cases of ear 
discharge. 

Tympanic membrane was not examined by Dr. Bronson 
unless there was reason to suspect an abnormal condition. 
She notes abnormalities (drum thickened, wax in canal, dull 
reflexes, etc.) in 19 cases, 6 girls and 13 boys. Mastoid normal 
for all girls, 2 boys have scars as a result of operation. She 
reports 3 cases of ear discharge. 

VISION 

Dr. Moore tested the vision of all his subjects, using the 
Snellen chart. No attempt was made to determine the na- 
ture of the visual defects. Owing to lack of suitable office 
room, vision tests of Dr. Bronson's subjects had to be 



226 TRAITS OF GIFTED CHILDREN 

omitted. Below is a summary of the findings of Dr. Moore. 
In this summary, where the vision was unequal in the two 
eyes, the mean is used. There were 15 such cases among the 
boys and 21 among the girls. 

TABLE 71 
VISUAL ACUITY 

Dr. Moore* s subjects 
Boys (139) <Hrls (142) 

Visual acuity N % N % 

Normal, better than 10/15 106 76.2% 107 75.3% 

10/15 9 6.5% 18 12.7% 

10/20 9 6.5% 5 ' 3.5% 

Total, 10/15 or less 33 23.7% 31 21.8% 

Total, 10/20 or less 24 17.3% 13 9.2% 

Total, 10/30 or less 15 10.8% 7 4.9% 

Total, 10/100 or less 2 1.4% 1 0.7% 

Of the 64 children with vision 10/15 or less (33 boys and 
31 girls), 26 wear adequately corrective glasses. Several 
others wear glasses which do not correct, or have glasses 
which they do not wear. 

The school reported 20 per cent of gifted group and 16 
per cent of control group as having "subnormal'* vision. 

EYE DISORDERS 

Both physicians regularly made note of eye conditions 
other than vision. Here again the personal equation of the 
examiner enters, as Dr. Bronson 1 reports four and a half 
times as many abnormal conditions as Dr. Moore. 2 In 
the summary given in Table 72, the sexes have been 
combined. 

*Dr. Bronson: 

Blepharitis, a slight crusting ' of the eyelids is very common in 

children, especially in those tinder par or who have refractive error. 

Dilatation of the pupils during examination with light in the face is 

another in which the cause was organic disease. 
Strabismus, this is definite and the figures should be correct. 

2 Dr. Moore: 

Blepharitis, not reported unless causing symptoms. 
Strabismus, all cases reported showing any apparent deviation. 
Conjunctival injection, very mild cases so common they were not 
reported. 



MEDICAL EXAMINATIONS 



TABLE 72 
EYE DEFECTS OTHER THAN VISION 



227 



Dr. Moore 


Dr. Bronson 




N 


Per cent 


N 


Per cent 


Blepharitis 


1 


0.4% 


24 


7.8% 


Conjunctivitis 


10 


3.6% 


11 


3.6% 


Nystagmus 


1 


0.4% 


6 


1.9% 


Persistent dilatation of pupils 








14 


4.5% 


Styes 








4 


1.3% 


Strabismus 


2 


0.7% 


7 


2.3% 


Squint 








2 


0.6% 


Granular lids 


1 


0.4% 


1 


0.3% 


Chronic catarrhal condition 








2 


0.6% 


Poor muscle control 








1 


0.3% 


Scleritis 








1 


0.3% 


Von Graefe's disease 








1 


0.3% 


Purulent discharge (slight) 








2 


0.6% 


Ptosis 


2 


0.7% 









Total 



17 



LIPS, GUMS, AND TONGUE 



76 



Dr. Moore reports no abnormalities of lips, gums, or 
tongue. 

Dr. Bronson reports lips normal in 78 per cent of her 
cases and "upper lip short" in 19 per cent. Gums are re- 
corded as normal in 75 per cent of all her cases. Gingivitis 
is reported for 12 per cent, "spongy gums" for 5 per cent. 
In 8 per cent of her cases the gums are described as un- 
healthy," "sore," etc. One 12 year old boy is reported as 
having receding gums. 

Abnormal conditions of the tongue are reported for 8 
cases. These are: large papillae, 3; fissured, 2; coated, 2; 
large tongue, one. 

SOFT PALATE 

Dr. Moore reports no abnormalities for the girls of his 
group. In the case of one boy the soft palate is much elon- 
gated, in another narrow, and in another the uvula is miss- 
ing. 

Dr. Bronson describes the soft palate as "high and 
narrow" in 8 cases (4 girls, 4 boys) ; as "high" in 3 cases 
(one girl, 2 boys) ; as "broad" in 2 cases (one girl, one boy) . 
One 10 year old boy has an extra uvula. 



228 TRAITS OF GIFTED CHILDREN 

HARD PALATE 

Dr. Moore reports only one abnormality, this a high arch 
for an 11 year old boy. 

Dr. Bronson reports a total of 116 abnormalities (47 girls 
and 69 boys) , as follows : 

Boys Girls Total 



Palate high and narrow 


37 


20 


57 


Palate narrow 


6 


8 


14 


Palate high 


22 


18 


40 


Palate broad and high 





1 


1 


Palate broad and flai. 


1 





1 


Palate broad 


2 





2 


Palate, no description 


1 





1 


Total abnormalities 


69 


47 


116 


Per cent of total group 


38.8 


35.6 


37.4 



The above figures show that when two exceptionally 
competent and well-trained physicians examine the hard 
palates of children, one may report many times as many 
abnormalities as the other. We have no reason to suppose 
that abnormalities of hard palate are actually more numer- 
ous in the bay cities of California than in Los Angeles. 

SECOND DENTITION 

The number of first teeth and permanent teeth was re- 
corded by both physicians for each child. The following 
figures show the mean numbers for each age. 



Mean No. of 

2, 3, 4 

5 

6 

7 

8 

9 

10 
11 
12 
13 
14 
15, 16 



TABLE 73 
PROGRESS OF SECOND DENTITION 

Dr. Moore Dr. Bronson 


Boys 


Girls 


Boys 


Girls 


1st 


2nd 


1st 


2nd 


1st 


2nd 


1st 


2nd 


20 


0. 


20.0 


0. 


20.0 


0. 


20.0 


0. 


18 


3.7 


20.0 


1.0 


19.0 


1.0 







13 


9.7 


18.6 


4.6 


17.5 


4.3 


18.0 


4.0 


15.1 


7.4 


12.9 


9.6 


8.3 


12.3 


13.8 


8.6 


12.6 


11.0 


10.5 


12.0 


11.9 


10.6 


10.6 


12.3 


10.8 


12.8 


9.3 


13.9 


10.3 


13.1 


7.5 


16.1 


9.4 


13.8 


6.4 


18.0 


6.3 


16.6 


5.5 


18.3 


4.5 


20.5 


2.3 


23.9 


4.2 


20.6 


1.8 


23.4 


1.9 


24.3 


2.1 


24.8 


0.7 


25.8 


0.7 


26.0 


0.5 


26.4 


0.2 


27.3 


2.0 


24.0 


0.3 


26.8 


0. 


27.2 


0. 


28.0 


0. 


27.0 


0.5 


25.8 


0. 


28.0 


0. 


28.0 


- 





0. 


25.0 



MEDICAL EXAMINATIONS 229 

It will be noted that the sex differences in mean number 
of permanent teeth are not very marked. In Dr. Moore's 
group, the boys have the larger number of permanent teeth 
before age 7, and the girls a significantly larger number at 
10 and 11. In Dr. Bronson's group, the boys have a slightly 
larger number below 8, and the girls a larger number at 9, 
10, and 11. 

CONDITION OF TEETH 

Our summary here is for Dr. Moore's group only, as Dr. 
Bronson omits report for 54 of her 310 cases and statistics 
on the remainder would therefore be ambiguous. 

TAJBLE 74 
CONDITION OF THE TEETH 

Per cent of children Mean number of Mean number of 

without cavities filled cavities unfilled cavities* 

Boys Girls Boys Girls Boys Girls 

2, 3, 4 100.0% 80.0% 5.0 

5 66.7% 100.0% 4.0 

6 0.0% 4Q.O% 4.0 0.3 0.3 3.0 

7 71.4% 58.3% 3.5 0.8 1.5 1.2 

8 11.1% 38.5% 1.5 2.4 1.4 1.4 

9 25.0% 27.3% 2.3 2.5 0.6 0.8 

10 33.3% 28.0% 1.4 1.3 1.1 1.7 

11 17.2% 40.0% 2.7 1.0 0.7 1.4 

12 25.9% 33.3% 2.1 2.3 1.1 0.8 

13 57.1% 23.0% 2.8 3.4 0,5 0.6 

14 20.0% 60.0% 2.0 3.0 1.0 2.5 
15, 16 0.0% 50.0% 3.0 5.0 0.0 0.0 

* For subjects having cavities. 

The dental conditions of these children are very much 
better than haVe usually been reported for unselected school 
children. This is probably due in part to better habits of 
personal hygiene and to better dental care. 

NOSE 

Nasal obstruction 1 and peculiarity of nasal form were 
noted by both physicians, with the results shown in Table 75. 

1 Dr. Moore : "Only nasal obstructions causing symptoms were reported. 
Slight obstructions due to spurs or to minor deviations of septum were not 
considered important enough to record." 

Dr. Bronson: "I think the proportion of nasal obstruction higher in 
the Bay region. I tested each nostril separately, and if child had to open 
the mouth to expire a deep inspiration, the case was counted as positive." 



230 TRAITS OF GIFTED CHILDREN 

TABLE 75 
FREQUENCY OF NASAL OBSTRUCTION 

Dr. Moore Dr. Bronson 



Alae-nasi (peculiarities of) 


1.3% 




Mouth breathing type 




9.6% 


Negroid type 




5.2% 


Deviation of septum 




0.4% 


Nostrils short, wide or narrow 


0.4% 


9.3% 


Obstruction 


3.9% 


17.0% 


, Discharge 


7.3% 


11.1% 


Excoriation of nares 




1.5% 



Total 12.9% 54.1% 

Note: Per cents are per cents of the total cases examined. 

PHARYNX AND TONSILS 

The two physicians have not recorded their examinations 
in the same manner. Dr. Bronson has reported the present 
condition of the pharyngeal cavity; Dr. Moore, the presence 
or absence of adenoids. Dr. Bronson does not, as a rule, re- 
port removal of adenoids. 

Dr, Moore reports satisfactory removal of adenoids for 
55.40 per cent of all boys, with imperfect removal in an addi- 
tional 3.6 per cent of the cases. He reports 36.7 per cent of 
the cases as having no adenoids; and 3.6 per cent with ade- 
noids which are large or pathological. He reports adenoids 
removed for 37.3 per cent of the girls, which includes one 
imperfect removal; none present in 57.0 per cent, and large 
or pathological adenoids in 5.6 per cent. 

Dr. Bronson's figures for condition of pharynx are as 
follows : 

Boys Girls 

Normal 36.5% 40.9% 

Granular 18.5% 21.2% 

Catarrhal 17.4% 9.9% 

Acutely inflamed 15.7% 18.2% 

Underlined without explanation 11.8% 9.9% 

The outstanding fact in Table 76 is the large number 
who have undergone tonsillectomy, more than 50 per cent 
of the entire group. In this case the physicians agree fairly 
closely. The school reports about 44 per cent, the Home 
Blank about 49 per cent. Probably not more than 10 per 
cent of unselected school children in most cities of the 
United States have had tonsils removed. 



MEDICAL EXAMINATIONS 231 

TABLE 76 
CONDITION OF TONSILS 



Removed 


Dr. Moore 
Boys Girls 


Dr. Bronson 
Boys Girls 


Satisfactorily 
Imperfectly 


56.1% 
1.4% 


38.0% 
0.7% 


43.3% 
12.4% 


46.2% 
12.1% 


Total removed 


57.5% 


38.7% 


55.7% 


58.3% 



Present 

Slightly enlarged but 

not pathological 2.2% 0.7% 4.5% 2.3% 

Decidedly enlarged or cryptic 1 30.2% 46.5% 11.2% 10.6% 

Definitely pathological 4.3% 4.2% 3.9% 3.8% 

Entirely normal 2 5.8% 9.9% 18.0% 18.2% 

*Not necessarily pathological, however. 

2 Of Dr. Bronson's cases, 6.8 per cent are underlined without explanation. 

A striking fact in the above figures is the small proportion 
of tonsils not removed which are entirely normal. 

CONDITION OF THYROID 

Conditions of the thyroid were reported by the physicians 
as follows: 

Dr. Moore Dr. Bronson 

Boys Girls Total Boys Girls Total 

Normal 88.5% 83.1% 85.6% 90.0% 66.6% 80.0% 

Slightly enlarged 4.3% 10.6% 7.6% 7.8% 23.5% 14.5% 
Hyperthyroid 7.2% 6.3% 6.8% 2.2% 10.0% 5.5% 

There were no cases of hyperthyroidism below 9 years, 
and only two Cases of slightly enlarged thyroid below 8 
years. The ages 10 to 13 show the highest incidence of thy- 
roid trouble. 

Dr. Bronson reports that army records show more than 
the usual amount of "adolescent" goitre in the Bay region, 
which is in harmony with the above figures. 

In view of the possible influence of thyroid activity upon 
mental development, it is interesting to compare the IQ's of 
the hyperthyroid cases with those of the gifted group as a 
whole. There were 14 boys and 24 girls reported as showing 
appreciable hyperthyroid symptoms. These gave a mean 
IQ of 156 for the boys, (S.D. 13.5), and 146 for the girls (S.D. 
18.9). Of the above 38 cases, 15 were recorded as true hyper- 



232 TRAITS OF GIFTED CHILDREN 

thyroidism. These gave a mean IQ of 157 (S,D. 13.2) . As the 
mean IQ for the entire gifted group is 152 (S.D. 10), the dif- 
ference between the hyperthyroid cases and the total group 
is not statistically significant. Of course, it is necessary to 
bear in mind the fact that our gifted group, all of whom are 
within the top one per cent of the general school population, 
represents a very narrow range of intelligence as compared 
with unselected children, and that this might very well mask 
a true correlation of appreciable amount between intelli- 
gence and degree of thyroid activity. However, if the corre- 
lation were very large we would expect to find the percent- 
age of hyperthyroid cases in our gifted group larger than it 
is. Comparative data on the frequency of hyperthyroidism 
among the general child population of the cities covered in 
the survey are, unfortunately, not available. Both Dr. Moore 
and Dr. Bronson are inclined to believe that it is probably 
somewhat lower than in our gifted group. 

The expectation that the incidence of hyperthyroidism 
should be higher among the gifted than in the general popu- 
lation rests upon two facts: (1) the well-known connection 
between cretinism and thyroid deficiency; (2) the supposed 
relationship between thyroid activity and the production of 
the microsplanchnic and macrosplanchnic body types. The 
microsplanchnic type is believed to be associated with 
greater thyroid activity, and Naccarati finds that it also tends 
to be associated with greater intelligence. 1 It should be noted, 
however, that Naccarati claims but a small correlation be- 
tween morphologic index and intelligence. 

CONDITION OF THE CERVICAL GLANDS 

Conditions of the cervical glands were reported as fol- 
lows: 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

Glands normal 49.6% 61.3% 59.0% 57.6% 

Enlarged, tonsils present 23.0% 28.8% 16.9% 20.4% 

Enlarged, tonsils removed 27.4% 9.8% 21.9% 19.7% 

Enlarged, no statement 2.2% 2.3% 

Total enlarged 50.4% 38.7% 41.0% 42.4% 

*Sante Naccarati: The Morphological Aspect of Intelligence, Archives 
of Psychology, 1921, p. 44. 



MEDICAL EXAMINATIONS 233 

The incidence of enlarged cervical glands showed no 
marked or consistent age differences. It will be noted that 
the reports of the two physicians agree fairly closely. Cornell 
makes the following statement regarding the incidence of en- 
larged cervical glands among the general school population : 
"Because of the great frequency of adenoids and decayed 
teeth in childhood the secondary effect, cervical adenitis, is 
correspondingly frequent, about three-fourths of young 
school children possessing palpable small glands or 'ker- 
nels'." 1 The proportion in our gifted group is a little over 40 
per cent. 

CHEST DEFORMITIES 

No statistics are more unsatisfactory than those that re- 
late to the frequency of growth defects. One school physician 
may report several times as many as another in examining 
children of the same school population. The reason, of 
course, is the indefinite terminology used. Deviations from 
the ideal human form occur in every degree, and what de- 

TABLE 77 
CHEST DEFORMITIES 

Dr. Moore Dr. Bronson 

Boys Girls Total Boys Girls Total 

Beading 14 7 21 67 39 106 

Harrison's groove 27 6 33 91 58 149 

Pigeon Chest 9 2 11 27 14 41 

Shoemaker's breast 1 1 56 29 85 

Miscellaneous 1 2 3 21 21 42 

Total deformities 52 17 69 262 161 423 

Duplications more than 

one condition listed for 

the same child 20 3 23 134 77 211 

Total cases with 

deformities 32 14 46 128 84 212 

Per cent of group 23.0% 9.9% 16.4% 71.9% 63.6% 68.4% 

Total cases normal 107 128 235 56 46 102 

Per cent of group 77.0% 90.1% 83.6% 28.1% 36.4% 31.6% 

Mean number of de- 
formities per child 0.37 0.12 0.24 1.47 1.22 1.36 

^r. Walter S. Cornell: The Health and Medical Inspection of School 
Children, 1912, p. 286. 



234 TRAITS OF GIFTED CHILDREN 

gree shall be taken as worthy of notice is largely a matter of 
subjective judgment. In Table 77 it will be seen that Dr. 
Moore reports 83.6 per cent of his cases as free from de- 
formities of the chest, Dr. Bronson only 31.6 per cent. Dr. 
Moore reports an average of .24 deformities per child, Dr. 
Bronson 1.36. 

Chest expansion was not measured. The intercostal angle 
was checked by Dr. Moore as normal in all cases. Dr. Bron- 
son reports intercostal angle as follows : 

Boys Girls 

Obtuse 36.5% 50.0% 

Acute 11.8% 15.9% 

Right 50.6% 31.8% 

No report 1.1% 2.3% 

RESPIRATION RATE 

This was taken by Dr. Moore with child in sitting posi- 
tion; by Dr. Bronson with child in recumbent position. 
Both physicians took the record near the middle of the ex- 
amination hour. The means and standard deviations by age 
are given below. 

TABLE 78 
MEAN RESPIRATION RATE BY AGE AND SEX 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

M S. D. M S. D, M S. D. M S. D. 

2-3-4 21.0 1.0 24.6 4.5 23.0 27.0 

5 23.0 2.8 23.0 0.0 25.0 1.1 

6 19.6 1.8 21.8 .8 24.6 .7 25.0 6.5 

7 22,1 2.1 21.5 2.5 23.0 .6 23.8 0.9 

8 19.2 1.3 22.1 3.3 22.0 2.4 23.8 2.3 

9 19.8 2.4 22.0 3.2 22.2 3.0 22.5 2.9 

10 20.1 1.5 18.2 1.5 20.9 2.4 22.5 2.0 

11 18.8 1.4 18.6 1.7 21.8 2.5 20.9 2.3 

12 18.7 1.7 19.2 1.3 20.5 2.2 19.9 2.2 

13 18.2 1.9 18.7 2.5 20.6 2.4 20.1 2.9 

14 18.6 1.4 18.2 .8 20.5 2.5 17.5 0.8 
15-16 17.0 17.0 1.0 19.0 

Dr. Moore's records for each sex run lower than those of 
Dr. Bronson at almost every age. This holds both for the 
mean and for the standard deviation. Since Dr. Moore's 



MEDICAL EXAMINATIONS 235 

records were taken with child sitting, and Dr. Bronson's with 
child lying down, one would have expected a difference in 
the opposite direction from that which was found. 1 It is pos- 
sible that one examiner may arouse more excitement in the 
child than would be caused by another physician. 

Table 78 shows a marked decrease in respiration rate 
with age, but no very consistent sex differences. 

LUNG CONDITIONS 

Dr. Moore reports one 11 year old girl with possible tuber- 
cular lesion, and 7 boys with abnormal percussion sounds. 
Of the latter group, he reports 2 as possibly due to chest de- 
formity, one as due to an old pleurisy (negative to tuberculin 
test), another in which X-ray showed mottling and peri- 
bronchial thickening with some calcified glands (negative to 
tuberculin test), one with asthmatic condition, one which 
may be a case of latent tuberculosis, and one in which X-ray 
showed inactive tuberculosis. In another case X-ray of the 
chest showed some calcification and enlarged bronchial 
glands, so that there is a possibility of latent tuberculosis, al- 
though percussion sounds were normal and bovine tests were 
negative. Of his total cases, 96.8 per cent are checked as 
normal for all lung conditions. 

Dr. Bronson reports rales or rhonchi for 8 girls and 21 
boys. She finds no tubercular conditions, but reports one 
boy and one girl with generalized bronchitis, one girl and 
one boy as asthmatic, one girl with acute bronchitis, and one 
boy with bronchial asthma. Of her cases, 89.4 per cent are 
checked as normal for all lung conditions. 

HEART 

Dr. Moore reports 3 cases of functional cardiac murmur 
among boys, and 3 among the girls. He reports among boys 
4 cases of abnormal heart rhythm (2 of arythmia, one extra 
systole, and one respiratory rhythm) ; among girls one case 
of arythmia and one in which the first sound is not clear. He 

*Dr. Moore: "The count was usually for 30 seconds, and was often re- 
peated." 

Dr. Bronson: "Count taken for one-half minute, without patient's 
knowledge, and unless especially rapid not repeated.** 



236 TRAITS OF GIFTED CHILDREN 

records 95.7 per cent of all his cases as having normal heart 
conditions (exclusive of heart rate), and he reports no cases 
of serious heart trouble. 

Dr. Bronson reports one boy and 4 girls as having heart 
affections of probably congenital origin; one boy and one 
girl as having had heart affected by rheumatic fever (the 
girl is said to be well at present, except for a slight dis- 
turbance of the heart rhythm) ; one boy and 2 girls as having 
acquired heart disease from other causes; and 20 boys and 
35 girls as having functional cardiac murmur. She reports 
disturbances of the heart rhythm for 17 boys and 17 girls. 
For all of these cases except two (both girls), the report is 
"respiratory rhythm/' which she considers has a probable 
vasomotor origin. The other two cases are arythmia, one, 
and extra systole, one. She reports "very rapid" heart beat 
(128) for one girl, "first sounds rather poor*' for 3 boys. She 
records 78.6 per cent of her boys and 59.8 per cent of her 
girls as normal for all heart conditions (exclusive of heart 
rate). In all, Dr. Bronson reports more than five times as 
many deviations from the normal for the heart as are re- 
ported by Dr. Moore. 

PULSE RATE 

The pulse rate was taken just before or just after respira- 
tion rate; by Dr. Moore with child sitting, by Dr. Bronson 
with child lying down. The distributions are irregular and 
the variability very high. For this reason the median and 
10-90 percentile range are used instead of the mean and 
sigma. The range is not given for the ages 2 to 6 and 14 to 
16, owing to the small number of cases at those ages. 

TABLE 79 
PULSE RATE BY AGE AND SEX 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

10-90 10-90 10-90 10-90 

Percentile Percentile Percentile Percentile 

Age range Median range Median range Median range Median 

2-3-4 92 92 93 93 

5 100 98 98 

6 89 97 90 93 

7 91 89 89 93 

8 81 76-98 85 79-97 89 78-98 90 



MEDICAL EXAMINATIONS 



237 



TABLE 79 Concluded 



Dr. Moore 

Boys Girls 

10-90 10-90 

Percentile Percentile 



Dr. Bronson 
Boys Girls 

10-90 10-90 



Percentile Percentile Percentile Percentile 

Age range Median range Median range Median range Median 



9 


72-91 


82 


78-95 


89 


10 


73-92 


81 


74-97 


87 


11 


70-95 


82 


73-96 


79 


12 


68-91 


77 


72-96 


85 


13 


70-91 


84 


74-91 


78 


14 




77 




74 


15-16 




73 




78 



84-116 


86 


84-110 


90 


70-96 


83 


84-105 


86 


78-93 


85 


73-98 


86 


74-92 


85 


78-94 


85 


73-92 


84 


73-118 


82 




70 




77 









79 



Table 79 indicates that there is a reduction of 12 or 15 in 
pulse rate from 6 to 14 years and that there are no marked 
or consistent sex differences. In general the figures agree 
fairly well with those for unselected children. 

D'ESPINE SIGN 

The D'Espine sign refers to conditions of the bronchial 
glands. When subject with enlarged or calcified bronchial 
glands pronounces the vowel e, the sound is audible at lower 
vertebral positions than when the glands are normal. The 
stethoscope is used. 

Physicians do not always agree on the limits of normality 
with respect to the D'Espine sign. Dr. Moore considers a 
D'Espine below the third dorsal as of possible significance. 
Dr. Bronson considers the condition normal if the D'Espine 
sign is not below the fourth dorsal; if at the fifth or sixth 
dorsal she considers the glands possibly enlarged; if at the 
seventh or below, definitely enlarged. Other physicians 
whom we have questioned have shown similar differences 
of opinion. 

The following summary of findings for the gifted group 
shows a marked sex difference in favor of the girls. No con- 
sistent age differences were found. Comparative figures for 
unselected children are not available. 



Dr. Moore 
Boys Girls 



Dr. Bronson 
Boys Girls 



Not below fourth dorsal 
Fifth or sixth dorsal 
Below sixth dorsal 
Not stated or indefinite 



58.0% 
26.2% 
14.4% 

1.4% 



68.4% 

20.4% 

9.9% 

1.4% 



65.7% 

21.9% 

5.6% 

6.8% 



78.7% 

12.9% 

4.6% 

3.8% 



238 TRAITS OF GIFTED CHILDREN 

ABDOMEN 

Dr. Moore 1 reports 89.2 per cent of his boys and 87.3 per 
cent of his girls as entirely normal so far as abdominal con- 
ditions are concerned. This includes 5 cases in which the ap- 
pendix has been removed and recovery was normal. For 14 
girls and 10 boys he reports the liver as more than 1% cm. 
below costal margin. This condition is probably of little 
significance, since in no case was the liver more than 2 or 
3 cm. below the costal margin. Three boys and one girl have 
tenderness over McBurney's point; one girl has pin worms in 
the rectum; one girl and one boy fecal masses in the colon, 
and one boy small external hemorrhoids. 

Dr. Bronson 2 reports only 46.6 per cent of her boys and 
51.5 per cent of her girls normal as regards abdominal condi- 
tions. For the remaining 159 cases she records a total of 265 
abnormal conditions, relatively five times as many as are 
reported by Dr. Moore. 3 These are classified as follows : 





Girls 


Boys 


Total 




(N = 178) 


(N ~ 132) 


(N = 310) 


Distention : Above umbilicus 


4 


14 


18 


Below 


3 





3 


Region not stated 


23 


46 


69 


Recti: Diastasis 


15 


31 


46 


Other abnormalities of recti 


12 


41 


53 


Fecal masses 


7 


.12 


19 


Other abnormal conditions of intestines 


1 


6 


7 


Symptoms of chronic appendicitis 


1 


1 


2 


Hernia 8 at present 


7 


12 


19 


Hernia in infancy (umbilical) 


4 


6 


10 


Miscellaneous minor conditions 


8 


7 


15 


Liver more than 1H cm. below 








costal margin 


1 


3 


4 


Total 


86 


179 


265 



Dr. Bronson reports appendix removed (with normal re- 
covery) for 4 boys and 5 girls. 

*Dr. Moore: "Examination on table with patient relaxed. Minor disten- 
tion not considered. Separation of recti not reported unlesg very apparent." 

*DT. Bronson: "Slight diastasis of the recti with small hernial open- 
ing above or at umbilicus is very frequent. The 'postural* child usually has 
prominent abdomen. The standard I tried to follow was that used with 
Harvard students." 

*Dr. Moore has reported hernia under genitals. He reports four cases; 
2 of inguinal-scrotal and 2 scars from former hernia. 



MEDICAL EXAMINATIONS 239 

GENITALS 

Dr. Moore reports 94.2 per cent of his boys, and 38 per 
cent of the girls who were examined, as normal in all re- 
spects so far as the genitalia are concerned. However, ex- 
amination of the girls was made in only 46 per cent of the 
cases. Of those examined abnormal conditions were found 
in 11 cases, or 16.9 per cent, but this should not be taken as 
representative, as examination was made in many cases 
because of suspected abnormal condition. The 11 cases all 
showed vaginal discharge as the active symptom. Micro- 
scopic examination of the smear was made in 8 of these 
cases, with negative results in 6. In one case diphthyroids 
were present; in one, streptococcus and micrococcus ca- 
tarrhals. 

For the boys he reports one case of undescended testicles, 
one of tight foreskin, one of infantile penis (12 year old 
boy), one case in which the testicles were inclined to retract 
into the ring. He reports 2 cases of mguinal-scrotal hernia 
and 2 cases with scars from former hernia. Of the boys 62.3 
per cent have been circumcised. 

Dr. Bronson reports 50.5 per cent of her boys and 52.9 
per cent of her girls normal as regards genital conditions. 
Examination of the genitals was not made for 32.5 per cent 
of the girls, nor for 3.9 per cent of the boys. She reports 15 
cases of undescended testicles (including cases of testicles 
in canals), and 8 additional cases in which the testes are 
either very small or very large. The scrotum is described as 
"very small" in 9 cases and "very large" in 4 cases. In 8 
cases minor abnormalities of the penis are reported. In 15 
cases the word genitals is underlined without explanation. 
48.9 per cent of the boys have been circumcised. 

Dr. Bronson reports abnormal conditions of the genitals 
for 7 girls (a total of 12 symptoms reported). This is 7.9 per 
cent of the number examined. Conditions were as follows : 

Irritated condition of external genitalia 7 

Large clitoris (hymen intact) 2 

Circumcised because of smegma 1 

Discharge, probably gonococcal 1 1 

Swelling of external genitalia 1 

l Parents were advised to take this child to the family physician for treatment. 



240 TRAITS OF GIFTED CHILDREN 

SPINAL DEFORMITIES 
Spinal deformities and defects of posture were reported 

as shown below : 

TABLE 80 

DEFECTS OF POSTURE 

Dr. Moore J>r. Bronson 

Boys Girls Boys Girls 

Kyphosis: . __ 

MoSrate 1<U% 4.2* 3^% sio* 

Marked 6.5% 2.1% 0.0% 0.0% 

LOF S Sligkt 0.7% 0.7% 3.4% 6.1% 

Moderate 1.4% 2.1% 8.4% 11.4% 

Marked 2.2% 0.0% 1.1% 2.3% 

C01 S Slight 4.3% 3.5% 6,7% 2.3% 

Moderate 2.9% 0.7% 1.7% 1.5% 

Marked 0.0% 0.7% 0.0% 0.0% 

Fatigue posture : ^ A ^ _ 

Slight 5.8% 2.8% 18.0% 12.1% 

Moderate 9.4% 7.1% 15.7% 14.4% 

Marked 3.6% 2.1% 12.3% 7.6% 

Other abnormalities 3.9% 8.3% 

Entirely normal 42.5% 61.2% 30.4% 33.3% 

It seems that the differences between the reports of the 
two physicians are here due in part to a difference in no- 
menclature. Cases of the kind which Dr. Bronson records as 
fatigue posture are by Dr. Moore recorded as slight kypho- 
sis. Dr. Moore reports somewhat more than Dr. Bronson as 
free from all kinds of spinal deformities. Both report only 
a small number of cases of "marked" defect. The literature 
of school hygiene contains innumerable reports on the per- 
centage of school children having spinal deformities, but the 
results show such wide disagreement as hardly to be worth 
quoting. One of the most careful studies, that of Scholder, 
Weith and Combe of 2,314 school children of Lausanne, 
Switzerland, gave 24.6 per cent with scoliosis and 5.8 per 
cent with kyphosis or lordosis. The Lausanne figure for 
scoliosis is considerably higher than that found for our 
gifted. 1 On the other hand, statistics on spinal deformities 

Beholder, Weith and Combe: Les deviations de la colonne vertebrae 
dans les coles de Lausanne. Jahrb. der schweizerischen Gesellshaft f. 
Schulhygiene, 1901. For other data on the incidence of growth disorders, 
see L. M. Terman, The Hygiene of the School Child, pp. 73 ff. 



MEDICAL EXAMINATIONS 241 

in colleges not infrequently show from 60 per cent to 80 
per cent with defects of posture. Lloyd T. Brown, for ex- 
ample, reports 80 per cent of Harvard freshmen as having 
"unsatisfactory" posture. 1 

EXTREMITIES 

Dr. Bronson recorded 17 per cent of boys and 9 per cent 
of girls as having bow legs or knock knees in "moderate" 
degree; and 3.3 per cent of boys and 3.8 per cent of girls in 
"marked" degree. Dr. Moore reported no cases under this 
heading. 

Dr. Moore records 10 per cent of boys and 12 per cent of 
girls as having either "beginning flat foot," "some pronation 
with low arches," or (three cases) "marked" flat foot. Dr. 
Bronson records 62 per cent of boys and 48 per cent of girls 
as having one of these conditions. However, pronation ac- 
counts for more than three-fourths of her positive cases. 
She records only 5 cases of "marked" flat foot in 310 sub- 
jects. 

Dr. Moore reports 6 other children with minor abnor- 
malities of the feet or legs, besides a serious case of arthritis 
deformans; Dr. Bronson, 22 additional cases, all minor ex- 
cept one showing after effects of infantile paralysis. 

Dr. Moore reports one boy and one girl whose hands 
show some trophic disturbances; one girl whose arm and 
hand and leg muscles are atrophied from arthritis deform- 
ans; and one girl whose arm and hand are paralyzed from 
infantile paralysis. Dr. Bronson reports 16 boys and 13 girls 
with minor deformities of the arms or hands, such as "in- 
curved fingers," "clubbed fingers," etc. 

Subtracting duplications (more than one condition re- 
ported for the same child) only 18.5 per cent of Dr. Bron- 
son's boys and 23.5 per cent of her girls are recorded en- 
tirely normal as regards extremities. Of Dr. Moore's cases, 
87 per cent of boys and 84.5 per cent of girls are recorded as 
normal. 2 

a Lee, Geer and Brown: Bodily Mechanics in Harvard Freshmen. Ameri- 
can Phys. Ed. Rev., November, 1920. 

2 Dr. Moore : "In the case of bow legs and knock knees, slight variations 
from the normal were not reported. Flat feet reported only when the 
arch was actually fallen." 



242 TRAITS OF GIFTED CHILDREN 

It should be noted that these classifications do not include 
vasomotor disturbances, all of which were tabulated under 
skin. 

BONES 

Dr. Moore 1 reports all boys as normal. He reports two 
girls (identical twins) as having deformities due to rickets, 
and three other girls with minor abnormalities of the bones. 
Dr. Bronson 2 reports 18.5 per cent of boys and 15.9 per cent 
of girls as having enlarged epiphyses or abnormalities of the 
diaphyses. The conditions here reported do not include 
those which have been reported under deformities of the 
chest, spine, extremities, etc. 

X-ray photographs were secured by Dr. Moore for 57 
gifted cases selected at random. The measurements of these 
by Dr. Baldwin have been presented in Chapter VII. 

JOINTS 

Dr. Moore' reports 100 per cent of his boys and 98.6 per 
cent of his girls as normal. Dr. Bronson 4 reports 32 per cent 
of her boys and 27,3 per cent of her girls as showing hypo- 
tonicity of joints. In addition she reports "loose elbow 
joints" for one boy, "loose knee joints probably due to dis- 
placed cartilage" for another, and "apparently no socket at 
right hip" for a third. One girl has the left femur out of 
socket, causing one inch shortening, and two have stiff 
finger joints. The word joints is underlined without explana- 
tion for 14 boys and 10 girls. 

MUSCLES AND TENDONS 
All of Dr. Moore's 5 boys and 97.9 per cent of his girls are 

a Dr. Moore : "Only diseases of the bones were considered." 

3 Dr. Bronson: "Bone defects reported by me were mostly enlargements 
of the ends of long bones. These usually occur in those having chest de- 
formities and slight knock knee or bow leg (more striking in severe 
rickets, of course)." 

*Dr. Moore: "The normal tonicity varies so widely that only patho- 
logical cases were reported." 

4 Dr. Bronson: "Hyp ertoni city is certainly very frequent but perhaps 
not pathological. This depends on one's ideal of the human body." 

*Dr. Moore: "Only atrophy and hypertrophy of pathological nature re- 
ported." 



MEDICAL EXAMINATIONS 243 

reported normal. Dr. Bronson 1 reports 70.2 per cent of her 
boys and 81.1 per cent of her girls as normal. Practically all 
the deviations from normal which she reports are minor 
conditions, described chiefly as "muscles not well devel- 
oped," etc. 

NEUROLOGIC CONDITIONS 

Neurologic conditions were reported as shown in Table 81. 

Here the differences between the two physicians are very 

great indeed. Nearly three times as many are reported en- 

TABLE 81 
NEUROLOGIC CONDITIONS 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

(N = 139) (N = 142) (N = 178) (N = 132) 

Tendon reflexes: 

Moderately exaggerated 1 23 48 43 

Markedly exaggerated - 12 17 

Sluggish or absent 1 1 

Tremor of spread fingers: 2 32 

Fine 62 48 

Coarse - 63 

Striking 86 

Tics or habit spasms 12 1 11 4 

Chvostek '1 1 

Mydriasis - 19 23 

Positive Romberg - 32 14 

Stuttering or Stammering - 32 

Coordination poor - 27 16 

Abnormal conditions recorded 17 57 228 177 

Entirely normal 71.2% 71.1% 28.6% 24.2% 

tirely normal by Dr. Moore as by Dr. Bronson. 8 However, the 
physicians agree in finding tics and habit spasms far more 
frequent among boys than among girls. This is supported 
by the home and school data. 

*Dr. Bronson: "It is hard to set a standard, but as a whole, my 'poor' 
cases will be those with diastasis of recti, hypotonicity, and other de- 
fects." 

3 Dr. Bronson: "Exaggerated reflexes, largely a matter of the personal 
equation. Tremor of spread fingers is present or not present; very many 
of us have it. 'Poor coordination* refers to inability to bring tips of the 
two forefingers together with the eyes shut. Mydriasis, pupils persistently 
dilated during the examination. These conditions can be present without 
organic disease. Most positives are in high-strung children, though not in- 
frequently a phlegmatic child may show poor coordination." 



244 TRAITS OF GIFTED CHILDREN 

Mental symptoms are reported as follows : 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

Excitable 33 35 29 

Irritable. 12 2 11 9 

Excitable and irritable 1 1 ' 

Recorded as normal 89% 96% 70% 65% 

ENDOCRINE SYMPTOMS 

Dr. Moore reports for boys 14 other cases of possible 
endocrine disturbances, such as overweight, tremor of 
spread fingers, etc., and for girls, 16 cases. Dr. Bronson re- 
ports 9 cases of possible endocrine disorder among boys, 
and 19 among girls. In all, 8.2 per cent of the gifted boys 
and 11.3 per cent of the gifted girls are reported as having 
endocrine symptoms. 

CONDITION OF URINE 

The procedure followed by the two physicians is indi- 
cated in the following statements. 

Dr. Moore: "Albumen test, with 3 per cent acetic acid; a 
few drops in cold urine, then gently heated to the boil to 
differentiate nuclear and serum albumen and the phos- 
phates. A few drops of the acid added during the heating 
process. Sugar test with Benedict's Solution. Indican test 
(not called for in examination schedule) with HCL, KCLO 3 
and chloroform." 

Dr. Bronson: "Two specimens, one brought from home; 
first, morning specimen; other, office specimen. Trace of al- 
bumen in office specimen might easily be less frequent, be- 
cause a rather large proportion of my patients were exam- 
ined in the morning, except on Saturdays. The procedure 
followed in testing was as follows : 

"1. Heat urine to boiling look for cloudiness. 
"2. Add about % to 1 c.c. of dilute 5 per cent acetic acid. 
Heat again and look for cloudiness in heated portion. 

"Albumen in p.m. specimens does not mean kidney dis- 
ease. It most frequently occurs in cases of extreme lordosis 
and is due to interference with the circulation in the left 
kidney only." 



MEDICAL EXAMINATIONS 245 

TABLE 82 
RESULTS OF URINE TESTS 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

Normal 69.0% 65.5% 83.1% 74.2% 

Orthostatic 13.7% 16.9% 14.6% 16.7% 

Sugar 0.0% 0.7% Not tested 
Indicanuria : 

Slight 2.9% 4.2% 

Moderate 10.8% 4.9% 

Marked or extreme 5.7% 12.0% 

Other conditions 0.0% 4.9% 2.2% 9.1% 

Under "other conditions" Dr. Moore reports the following cases, 
all among girls: 

1 case true alhumenuria with granular cast 
1, slight true albumenuria 

1, true albumenuria 

2, excess pus cells 
1, few hyalin casts 

1, numerous granular casts. 

Under "other conditions" Dr. Bronson reports as follows: 

For boys : 

1, heavy trace of albumen, acid 1018, few white blood cor- 
puscles 

1, slight cloud with sediment, both specimens 
1, question of true albumenuria 
1, acid, Turbid 1010, pus and epithelial cells. 

For girls: 

1, albumen in afternoon specimen and pear-shaped epithelial 
cells 

1, probably true albumenuria with urates 

1, albumen and pus cells in afternoon specimen 

1, 1 to 2 white blood corpuscles to each field 

I, heavy albumen in afternoon, little in morning; pus and epi- 
thelial cells 

1, both specimens albumen with pus 

1, albumen in afternoon, and acid 1012 

1, clump of pus cells, a little albumen sediment; later test 
showed no albumen after removal of pus 

1, albumen in afternoon and square epithelial cells 

1, 1 to 2 pus cells to each high power field; trace of albumen and 
acid in both specimens 

1, light cloud of albumen afternoon and morning, much mucus 
and square epithelial cells, few white blood corpuscles 

1, square epithelial cells. 



246 TRAITS OF GIFTED CHILDREN 

HEMOGLOBIN 

The Talquist test was used. Procedure is indicated in the 
following comments. 

Dr. Moore : "The book used was one patented June, 1902, 
and sold by Edw. Pennock, Philadelphia. The second drop 
was used and read as soon as dry, usually in 20 or 30 
seconds." 

Dr. Bronson : "Alas ! there are several firms putting out 
Talquists. Mine is the authorized, signed Talquist. It is a 
color reading, and even in the acid-hematin method, which 
takes a lot of blood and time, the standard in hospitals is 
always varying." 

The lack of objectivity in the hemoglobin test is indi- 
cated in Table 83, in which it will be noted that Dr. Moore 
reports only about 1 per cent as below 75, as compared with 
approximately 25 per cent reported by Dr. Bronson. 

Dr. Moore's records were made in such figures as 79, 83, 
etc.; Dr. Bronson's in round numbers by intervals of ten, 
as 60, 70, 80, etc. In order to allow comparisons, Dr. Moore's 
data have also been grouped in intervals of ten. 

TABLE 83 
RESULTS OF HEMOGLOBIN TESTS 

Dr. Moore Dr. Bronson 

Boys Girls Boys Girls 

Below 65 0.0% 0.0% 0.6% 0.8% 

65-75 1.4% 0.7% 23.6% 26.5% 

75-85 70.5% 76.1% 64.6% 61.4% 

85 or above 26.6% 22.5% 6.7% 8.3% 

Not examined 1.4% 0.7% 4.3% 3.0% 

BLOOD PRESSURE 

Dr. Moore: "The first half of the cases were done with 
arm at side, both lying and standing. The last half with 
arms extended, both lying and standing. In the former, I 
often found a rise in pressure on standing, in the latter 
usually a fall. 59 

Dr. Bronson: "I took the pressure until constant, with 
child lying down. Then child got off the table and I took the 
pressure again quickly, this time with patient's arm ex- 
tended toward me or on my knee (in the case of youngest 



MEDICAL EXAMINATIONS 247 





P><N jl lO lO O 

T-( 2; TH TH 

8 "^ 
.^ S cooo 


O ITS 

m co oo 

TH 
TH OO 




OO CM 






8 ^ || ^ O ^ 






e~3 <^ 






& ^0> 5r CO CO TH 

B S3 o ^ 

t-, TH 


CO ^ CM 

O TH 
TH 




CO t-C "^ 






a ^co 







t-^ co 


v5 












TJ 




>itn " J2 ^ CO 


TH ^ O 




TH iri co 


^ 

Q 




* OS 


OS TH 






O 




S .ggg cor. 


CO O 


t- ^ 


O CM 







&J. 11 00^^ 

*g iz comco 

** TH TH 


CM TH O 

TH <rH 


TH 00 CO 
TH TH 


co ^ co 


.S 




TH 


TH 


TH 




el 













cd 




Q . ,H CD CM CO 


OS CO 


CM CO 


TH LO 


C/3 




^ ^ rH || ^^ TH CO 




^, CO CO 


CO ^ ? 


_j 




f*Oi > CO OS OS 

<~^ U ** O 
* O rH 


CM CO O 

O TH 
^4 


M* TH TH 
TH TH 


co irS co 


a 

01 












to 













G 




A -~. CO ^* ^S 


CO lO 


O Tf 


TH CM 






1 11 r*ii "^* ^" 




_^ QJ ^O 




K*S 




^J *>"S ^ CO O OO 


TH O 


TH OS OO 


CO CO CO 


I-H 




^ ^ 


TH 
TH 


OS 




c 

QJ 


Tt* 

OO 


*<< j 22 os -"f 

^ 2* w" 2 ^- 

l^ 1 ^^ ^, T^ TH CO 


!> TH 
^ TH CM* 




^ CO CO 


"S 


S 


^ ^ TH TH 
TH 


TH TH 






Q 
C 




d 










H 


o m 
p 55 . ^ o 10 co 
co gS^^ ^osco 


*< CM 




** OS 


(D 




gS ^>>osll OlOOO 


0^0 




O CO CO 


|tt 




S W x TH o 


TH O TH 
YH 




TH 


TJ 




^ (3 













Q . r*. 

g M 00 W TH l>- 


CO CO 




TH CM 


*- 






j.^ Irt OO 




^ 1> OS 


t5 




i-3 55 CM O5 OS 


CM t^ OO 




CM 10 CM 


flj 






OS 






a> 




cT U5 O5 








tn 




tt 7 ^ ^So 


CO CO 
OO ^"* ^ 


CO I> 


CO CM 

OS^^. 


a; 
o 




^ jz ^f co os 


TH O 


oo r-I os 


^ ^ CO 


c 




TH 


TH TH 


TH 









TH 








O) 




03 10 
rl rH CO TH C** 


TH -^ 


lO CM 


<*< 


ctt 

iH 




S 7 n iw^: 


rH OO TH 


Tf ^ ^ 


^ 1C iO 


Q 




g ^05 2 CO OS OS 


CM ifi CM* 


Tf O OS 


CO ^ CO 






^ 


TH 


TH 




^> 




T ( 








"o 




o! os oo 

g f CM ^ ^ 
>0 2 CM CO OO 


CO 
kO CO 

CM" ^ 


CO OO 
O 

TH o CM 


CO TH 

cM^: ^ 

CM Tf< CM 


: 




* 


- ^ 


r- OTH 








TH 


c; 


S ^ 











* 


a 


V 






O 


^^j 


5$ <u 


^ 





? J^ 00 5^> 5^ SQ 



248 TRAITS OF GIFTED CHILDREN 

patients who did not hold arm steady). I think the effect of 
position is of little significance as a test of vasomotor stabil- 
ity. A few seconds in getting the second reading makes 
much difference, especially in children." 

The blood pressure records are summarized in Table 84. 

Table 84 shows no marked or consistent differences 
between mean records found by the two physicians. The 
only consistent sex difference is the tendency of girls to a 
slightly higher average blood pressure at the ages 12 to 15. 
The increase with age is marked, as would be expected. In 
the records of Dr. Moore's boys, recumbent position, the 
(Pearson) correlation of blood pressure with age was .41. 
In four comparisons out of ten (Table 84) the S. D. is sig- 
nificantly greater for the age group 12 to 15 than for the age 
group 9 to 12, and in only one case significantly lower. 

The following comparative data, for girls, are taken from 
Burlage r 1 

TABLE 85 

BLOOD PRESSURE NORMS FOR ITHACA SCHOOL GIRLS 
(Systolic Readings. Sitting Position) 

Means for Gifted Girls 
Age Cases Means (m.m.) (Recumbent position) 

Dr. Moore Dr. Bronson 

9 65 104.85 

109.12 106.06 



115.16 115.04 



10 77 108.69 

11 122 109.48 

12 92 114.16 

13 100 118.55 

14 90 123.71 

15 116 123.29 

16 75 118.92 

17 121 115.41 

18 175 111.65 

Note. The averages for our gifted girls of 9 to 12 years, and 
12 to 15 years, are added for comparison. 

Blood pressure records are likely to be Considerably af- 
fected by the emotional factor. Excitement may increase the 
readings by as much as 30 m.m. The records of Burlage were 
apparently obtained under very favorable conditions. The 
nurse and school physician who took the records were 

Stanley R. Burlage: The Blood Pressure and Heart Rate of Girls dur- 
ing Adolescence. Amer. J. of Physiology, April, 1923. 



MEDICAL EXAMINATIONS 249 

known to the children and it was felt that the emotional 
factor was reduced to a minimum. Burlage's records run a 
little higher than those for our gifted group for ages 12 to 15, 
but slightly lower for ages 9 to 12. The conclusion seems to 
be warranted that gifted children show no significant devia- 
tion from unselected children with respect to blood pressure. 

SUMMARY 

Medical examinations were given to 783 gifted children, 
of whom 591 belonged to the main experimental group. 
The data for these 591 cases have been analyzed, yielding 
the following results, among others : 

1. Tests show 6 cases (about 1 per cent) with "decidedly 
defective" hearing and approximately 10 per cent with 
hearing "somewhat defective." 

2. Vision tests show 10.8 per cent of boys and 4.9 per cent 
of girls with 10/30 vision or less. 

3. The mean number of permanent teeth present at eight 
years is approximately 11 for boys and 12 for girls. This is 
perhaps slightly better than normal. The mean number be- 
fore 8 years is higher for boys than for girls; after 8 it is 
appreciably higher for girls. 

4. Between the ages of 8 and 13 years one or more dental 
cavities (filled or unfilled) are found in the case of about 
three-fourths of the boys and two-thirds of the girls. In this 
age range the mean number of unfilled cavities per child is 
approximately one. 

5. Somewhat more than half of the entire group have 
undergone tonsillectomy. Only about a fourth of the remain- 
ing subjects have tonsils which are entirely normal. 

6. Hyperthyroidism is found in 6.1 per cent of cases. In 
this group there is no significant correlation between hyper- 
thyroidism and IQ. 

7. Approximately 43 per cent of the subjects have en- 
larged cervical glands. 

8. About one-half of one per cent showed symptoms of 
active or probably active tuberculosis. 

9. Mean respiration rate decreases from 21 at age 8 to 
18 at age 13, according to Mr. Moore's records. Dr. Bronson's 
records show a decrease from 23 to 20 in the same period. 

10. Mean pulse rate decreases from 83 at 8 years to 81 at 



250 TRAITS OF GIFTED CHILDREN 

13, according to Dr. Moore's records, and from 89 to 83 ac- 
cording to Dr. Bronson's. 

11. Dr. Moore records the D'Espine sign as present be- 
low the sixth dorsal (indicating enlargement of the bronchial 
glands) in 14.4 per cent of boys and 9.9 per cent of girls. Dr. 
Bronson's figures are 5,6 per cent for boys and 4.6 per cent 
for girls. 

12. Abnormalities of the genitals were rare. 

13. Urine tests gave results entirely normal in two-thirds 
of Dr. Moore's subjects and in three-fourths of Dr. Bron- 
son's. There was only one case of marked sugar content. In- 
dicanuria was present in approximately 9 per cent of the 
cases tested. 

14. Using the Talquist hemoglobin test, Dr. Moore finds 
about 99 per cent of his subjects above 75. The corresponding 
figure for Dr. Bronson is approximately 74 per cent. This 
difference may be partly accounted for by the greater 
amount of sunshine in Los Angeles, where Dr. Moore's sub- 
jects reside. 

15. Metabolism tests of 93 random cases, chiefly from 10 
to 13 years of age, showed 91.4 per cent normal or above. 

16. The blood pressure records agree closely at all ages 
with the norms of Burlage for unselected children in Ithaca, 
New York. For the ages 9 to 12 they are slightly higher than 
Burlage's and from 12 to 15 slightly lower. 

17. Marked kyphosis is reported for 2 per cent of the sub- 
jects, marked lordosis for 1.3 per cent, and marked scoliosis 
for but one subject. 

18. Comparison of the records of the two physicians 
shows that the personal equation enters largely into the re- 
ports upon skin conditions, eye conditions (other than vis- 
ion), conditions of the lips, gums, and tongue, abnormalities 
of the hard palate, deformities, heart disease, D'Espine sign, 
nervous symptoms, and hemoglobin. Respiration and pulse 
records are also subject to constant error due to the personal 
equation, but to a less degree. On some of the points men- 
tioned in this paragraph one highly competent examiner will 
report several times as many cases of defect as another ex- 
aminer of equal competence. Medical examination methods 
are less objective than those customarily employed by psy- 
chologists in psychometrics. 



MEDICAL "EXAMINATIONS 251 

19. Notwithstanding occasional disagreements in their 
results, the examining physicians are in accord in the belief 
that on the whole the children of this group are physically 
superior to unselected children of corresponding age in the 
school population. 

Dr. Moore: "In regard to a general comparison of this 
group with unselected children, it is my opinion that major 
and minor defects are much less common in the former. I do 
not have suitable figures on which to base a comparison as 
to the relative incidence of various defects, but I have a 
strong conviction that, other things being equal, there is a 
direct correlation between physical health and mentality in 
children when studied in groups. In my opinion the physi- 
cal superiority of the gifted group is indicated by the higher 
average of nutrition and by superior stability, physical and 
mental." 

Dr. Bronson: "The examinations of the gifted group 
were the most satisfactory of any series of examinations I 
have conducted. The quickness of these children in com- 
prehending what was desired of them in the various tests 
was a delight. As a whole there was unusual ability to con- 
centrate attention, and self-consciousness was less notice- 
able than in the average child. The home care, cleanliness, 
and health habits, such as diet, hours of sleep, etc., indicated 
superior intelligence on the part of the parents. There were, 
of course, exceptions to all these points. Physically, also, the 
gifted child ranked above the average child of the commu- 
nity. Interest in games and outdoor sports was, I should 
judge, about the same as that of the ordinary child. The 
greater number of the defects recorded in my reports were 
minor in degree and such as are found in all civilized 
peoples. If our standard were as strict as that which we 
apply to blooded stock, we would find physical perfection in 
the human race very rare after early infancy." 



CHAPTER X 

SCHOOL PROGRESS AND EDUCATIONAL HISTORY 1 
AGE-GRADE STATUS 

From Tables 86 and 88, showing age-grade status at 
the time the children were located, it will be seen that if we 
use the commonly accepted standard and call a child re- 
tarded who has reached the age of 8 years and is below the 
second grade, or 9 years and is below the third grade, etc., 
then there is not a single child in the gifted group of 616 who 
is retarded. Even if we use a standard which is stricter by a 
half year there is still only one retarded. If we call each 
child accelerated who has reached the low first grade and is 
not yet 6 years old, the high first and is not yet 6%, the low 
second and is not yet 7, etc., then 84.5 per cent of the boys 
and 82.5 per cent of the girls must be classified as accel- 
erated. The per cent of acceleration increases relatively lit- 
tle after the age of 7 years, which means that most of the 
extra gain is made before that age. 

Tables 87 and 89 show the status of the same 616 chil- 
dren in relation to mental age. Here the situation is strik- 
ingly reversed. If we now call a child retarded who has 
reached the mental age of 7% years and is below the second 
grade, 8% years and is below the third grade, etc., then all 
but 4 of the boys and all but 2 of the girls are retarded. 

Table 90 gives the distributions of chronological ages and 
mental ages by even ages and even grades for the sexes 
combined. In this case the lines indicate the somewhat more 
conservative norms of grade location used by Ayres, Strayer, 
and others. Even on this basis, all but 12 of the children are 
below the grade normal for mental age, while not one is re- 
tarded with respect to chronological age. It will be noted 
that the discrepancy between mean chronological age and 
mean mental age is 2.8 years in the first grade, and that by 
the fifth grade the discrepancy is increased to nearly 5 years. 

1 Written with the assistance of Florence L. Goodenough. 

253 



254 



TRAITS OF GIFTED CHILDREN 

TABLE 86 
AGE-GRADE DISTRIBUTION OF BOYS 



Age 


1L 1H 2L 2H 3L 3H 4L 


4H 


5L 


5H 


6L 


6H 


7L 


7H 8L 8H 9L 


9H 


5* 


1 


















5-6 


1 1 


















6 


2111 


















6-6 


13312 


















7 


116211 


















7-6 


2373 






1 












8 


1293 


3 
















8-6 


1 12 


6 


6 














9 


3 


5 


3 


4 


4 


1 








9-6 


1 1 


9 


15 


12 


2 


1 


1 






10 




1 


4 


13 


8 


6 


3 


1 




10-6 






1 


2 


12 


12 


4 


6 1 




11 










4 


9 


12 


8 3 1 


1 


11-6 










1 


3 


4 


10 4 1 




12 














2 


7662 




12-6 
















841 




13 
















1331 


1 


13-6 


{* 














2 2 




Total 


5 6 5 11 10 18 23 


24 


29 


32 


31 


32 


26 


33 25 17 6 


2 


TABLE 


87 














Mental 


MENTAL AGE AND GRADE DISTRIBUTION OF BOYS 


Age 


1L 1H 2L 2H 3L 3H 4L 


4H 


5L 


5H 


6L 


6H 


7L 


7H 8L 8H 9L 


9H 


7 


1 


















7-6 




















8 


1 1 


















8-6 


1 1 


















9 


221 1 


















9-6 


1111 


















10 


112 1 


















10-6 


1 22121 


















11 


4321 


















11-6 


222 


















12 


1124 


2 


1 














12-6 


,152 


6 








1 








13 


4 


3 


3 


2 


1 


1 








13-6 


3 4 


4 


6 


4 


1 










14 


2 1 


4 


5 


6 


3 










14-6 


3 


3 


7 


5 


1 


2 




1 




15 






2 


8 


3 


9 


3 


1 




15-6 




1 


2 


5 


8 


5 


6 


2 




16 




1 


3 


2 


7 


6 


4 


6 2 




16-6 










3 


4 


4 


9231 




17 










2 


2 


4 


6541 


1 


17-6 










1 


2 




3611 




18 










1 




4 


4341 




18-6 
















6 4 




19 
















11 2 


1 


19-6 














1 


1 




Total 


5 6 5 11 10 18 23 


24 


29 


32 


31 


32 


26 


33 25 17 6 


2 



EDUCATIONAL HISTORY 255 

TABLE 88 

AGE-GRADE DISTRIBUTION OF GIRLS 
Age 1L 1H 2L 2H 3L 3H 4L 4H 5L 5H 6L 6H 7L 7H 8L 8H 9L 9H 10L 



5 1 
















5-6 1 
















6 1 


421 














6-6 1 


2 42 


1 












7 


143 


1 


1 










7-6 


3 


3 


4 










8 


1 


6 


6 


611 








8-6 


9 


2 


3 


8 4 








9 




1 


2 


56411 








9-6 






1 


46534 








10 








3858 


2 






10-6 








476 


7 


3 




11 








6 7 


12 


4 1 


1 


11-6 








1 


7 


3 4 


3 1 


12 










2 


4 4 


4 1 


12-6 










2 


2 7 


621 


13 












1 


4 1 1 


13-6 














32 1 



Total 4 6 3 9 18 14 17 23 20 22 22 27 32 17 16 21 7 2 1 

TABLE 89 

Mental MENTAL AGE AND GRADE DISTRIBUTION OF GIRLS 

Age 1L 1H 2L 2H 3L 3H 4L 4H 5L BH 6L 6H 7L 7H 8L 8H 9L 9H 10L 

7 1 

7-6 1 

8 

8-612 

9 1312 



9-6 


2 1 














10 


1233 1 














10-6 


1 4 














11 


1943 














11-6 


132 


4 












12 


3 


2 


3 


1 








12-6 


4 1 




1 


1 








13 


2 2 


4 


6 




1 






13-6 


1 


4 


3 


3 


1 






14 


1 3 


3 


2 


4 


2 


2 




14-6 


1 


3 




2 


4 


4 




15 




2 


4 


7 


2 


4 


1 


15-6 




1 


1 


1 


4 


5 


4 3 


16 








1 


6 


4 


73 1 


16-6 










2 


4 


73121 


17 












3 


54541 


17-6 








2 




1 


12441 


18 














525331 


18-6 














117 1 


19 














1 1 


19-6 














1 



Total 4 6 3 9 18 14 17 23 20 22 22 27 32 17 16 21 7 2 



256 



TRAITS OF GIFTED CHILDREN 



TABLE 90 
SUMMARY OF GRADE LOCATION BY AGE AND MENTAL AGE 

GRADE 





1 


2 


8 


4 


5 


6 


7 


8 


9 


Age 


CA MA 


CA MA 


CA MA 


OA MA 


OA MA 


CA MA 


OA MA 


OA MA 


, CA MA 


5 


7 
8 
9 
10 
11 
12 
13 
14 
15 
16 
17 
18 
19 


5 


9 


5 
23 


9 

47 


1 
12 
55 


17 
64 


1 
26 
60 


1 
18 
45 


2 

7 
8 


15 
1 3 


16 
3 1 


6 
9 
3 


30 
2 3 


8 
13 
5 
1 


30 
1 3 


10 
26 
13 
5 
3 


35 


12 
20 
26 
21 
4 
1 


31 
1 


7 
27 
31 
30 
6 
2 


19 
2 


5 
18 
40 
36 
11 
1 


15 


1 
20 
43 
25 
16 
3 




11 
33 
33 
2 


2 
5 
5 
5 


Mean OA 
Mean MA 
Dlfler'nce 


6.3 
9.1 
2.8 yrs. 


7.3 
10.4 
3.1 yrs. 


8.1 
12.0 
3.9 yrs. 


8.8 
13.3 
4.5 yrs. 


9.7 
14.6 

4.9 yrs. 


10.4 
15.8 
5.4 yrs. 


11.5 
16.9 
5.4 yrs. 


12.4 
17.8 
5.4 yrs. 


12.8 
18.3 
5.5 yrs. 



Note. Lines show normal grade location 
used by Ayres, Strayer, and others. 



according to the standards 



EDUCATIONAL HISTORY 



257 



SCHOOL PROGRESS QUOTIENT 

By comparing the age of a child who is in a given grade 
with the average or median age of the school population in 
that grade, it is possible to compute a school progress quo- 
tient. This was done for all of our gifted children who were 
enrolled in school grades above the kindergarten at the 
time of the test. The standard of comparison used was the 
median ages of the children enrolled in various grades of 
Oakland in 1921-1922, at mid-term. These standard ages 
were as follows : 



Grade : 
Mdn. Age: 

Grade : 
Mdn. Age : 



LI 
6-5 



HI L2 
7-3 7-8 



H2 
8-2 



L3 
8-9 



H3 
9-2 



L4 H4 L5 H5 
9-10 10-3 10-10 11-5 



L6 H6 L7 H7 L8 H8* L9* H9* L10* H10* 
11-11 12-4 12-10 13-3 13-10 14-4 14-10 15-4 15-10 16-4 



* Projected norms. 



The progress quotient (P.Q.) of a given child is obtained 
by dividing the standard age for the child's grade by the 
child's age reckoned as of the middle of the term. The P.Q. 
distributions are given in Table 91. 



TABLE 91 
SCHOOL PROGRESS QUOTIENTS OF GIFTED CHILDREN 



P.Q. 


Boys 


Girls 


Total 


135-139 


4 


3 


7 


130-134 


6 


4 


10 


125-129 


14 


16 


30 


120-124 


34 


49 


83 


115-119 


81 


51 


132 


110-114 


109 


88 


197 


105-109 


64 


55 


119 


100-104 


19 


15 


34 


95- 99 


4 





4 


Total 


335 


281 


616 


Mean P.Q. 


113.8 


114.5 


114.1 


S.D. 


7.1 


7.1 


7.1 


Mean IQ (for comparison) 


149.2 


148.4 


148.5 


S.D. of IQ 


8.8 


9.4 





258 TRAITS OF GIFTED CHILDREN 

The average progress quotient of 114 means that the typi- 
cal gifted child is accelerated in school, as compared with 
unselected children, to an extent equal to 14 per cent of his 
age. This would amount to about one grade of acceleration 
at age 7, or two grades at age 14. It will be noted that the 
amount of acceleration of the gifted group reckoned thus is 
somewhat greater than that indicated by the net gain from 
skipping. (See page 273.) This difference is largely accounted 
for by the earlier entering age of gifted children as com- 
pared with normal. 

The typical gifted child is about 48 per cent of his age 
above the norm in intelligence and about 14 per cent of his 
age above the norm in grade location. The difference be- 
tween 48 per cent and 14 per cent is 34 per cent. From one 
point of view we might say that the typical gifted child is 
under-promoted to the extent of 34 per cent, or approxi- 
mately one-third, of his age. This would mean a retardation 
of three grades at age 9, and four grades at age 12. It is, of 
course, conceivable that considerations having to do with 
social training justify a good deal of under-promotion of 
gifted children, but as far as mere ability to accomplish is 
concerned, it will be shown (in Chapter XI) that some two- 
thirds of the under-promotion found with this group is un- 
justified. 

TEACHER'S RATINGS ON QUALITY OF SCHOOL WORK 

The School Blank which was filled out for the gifted, and 
also for a control group of unselected children of the ages 
8 to 13, asked for a rating of the child's work in each school 
subject. 1 It will be recalled that the children of this control 
group were selected from the schools attended by the gifted 
children, and on the arbitrary basis which required that 
they be of exactly normal age-grade status. The ratings were 
secured by the use of the instructions shown in the blank on 
page 259. 

Mean ratings were worked out for all ages, and for the 
sexes separately, to see whether there was any tendency for 
ratings to increase or decrease with successive years of 

Control Group A, described 00 pp. 177-178. 



EDUCATIONAL HISTORY 259 

SCHOOL RATING BLANK 

Compare this child's school work with that of the average child of the SAME 
SCHOOL GRADE, and rate its quality for all the subjects named below WHICH ABE 
BEING PURSUED. Rate also application and deportment. On the line before each 
subject, write the figure 1, >2, 3, 4, 5, 6, or 7. 

1 as Very superior to average child of the same grade. 

2 sa Superior to average child of the same grade. 

3 as High average. 

4 = Average. 

5 =s Low average. 

6 = Inferior to average child of the same grade. 

7 = Very inferior to average child of the same grade. 

ATTITUDE TOWARD SCHOOL FOREIGN LANGUAGE PRACTICAL SUBJECTS 

.....Application French Agriculture 

Deportment German Bookkeeping 

Latin Cooking 

ART, ETC. Spanish Manual training 

......Drawing (free hand) Mechanical drawing 

Modeling HISTORY Sewing 

......Music (instrumental) - Aacient or medieval Shop work 

......Music (singing) QyiMOTMtaenslw ......Typewriting 

Painting, water colors, United States history 

etc * MATHEMATICS - 

Algebra SCIENCE 

ENGLISH w Arithmetic """rS 7+ 

. Composition Geometry ......Chemistry 

Debating or speaking Trigonometry General science 

Dramatics -... * Geography 

Grammar PHYSICAL EDUCATION Nature study 

Literature Folk dancing Physical geography 

Penmanship Games and sports Physics 

Reading Military training Physiology or hygiene 

Spelling Physical training . Zoology 

Here, rate any other subjects the child is taking which are not named above 



chronological age. As no such tendency could be observed, 
ages were combined. However, only the 8 to 13 year old 
gifted children were included, as these were the only ages 
represented in the control group. Table 92 gives the age 
mean for each school subject for each sex and intelligence 
group. It will be remembered that the plan called for ratings 
ranging from 1 (high) to 7 (low), with 4 as representing 
work of average quality. 

The school subjects were then listed again in order of 
greatest difference between the mean ratings for the gifted 
and the control groups. (Table 93.) In the difference 
values an abrupt break will be noted after number 15 
(agriculture). The subjects preceding this are those which 
call primarily for abstract thought, or (as in the case of 
public speaking and dramatics) self-confidence, the ability 



260 TRAITS OF GIFTED CHILDREN 

to adapt oneself to changing circumstances, and quickness 
of mental processes. 

TABLE 92 
MEAN RATINGS ON SCHOOL WORK IN THE VARIOUS SUBJECTS 

Gifted Gifted Control Control All All 

Subject Boys Girls Boys Girls Gifted Control 

Drawing 3,76 3.42 4.07 3.72 3.62 3.87 

Modeling 3.54 3.33 3.62 3.66 3.48 3.64 

Instrumental music 3.07 2.73 3.46 3.31 2.89 3.38 

Singing 3.67 2.87 4.13 3.65 3.24 3.89 

Painting 3.85 3.54 3.94 3.68 3.71 3.80 

Composition 2.36 3.09 3.95 3.91 2.25 3.93 

Debating or speaking 2.00 2.09 3.84 3.94 2.04 3.88 

Dramatics 2.57 2.25 3.76 3.52 2.40 3.64 

Grammar 2.39 2.09 3.94 3.58 2.25 3.76 

Literature 1.98 1.79 3.64 3.46 1.90 3.56 

Penmanship 4.17 3.29 4.27 3.57 3.79 3.92 

Reading 2.02 1.79 3.15 3.38 1.92 3.26 

Spelling 2.28 1.94 3.58 3,19 2.13 3.39 

Ancient history 1.98 2.12 3.53 3.68 2.04 3.60 

Civics or citizenship 2.18 2.05 3.41 3.62 2.12 3.49 

U. S. History 1.90 2.36 3.72 4.20 2.11 3.92 

Arithmetic 2.53 2.51 3.83 3.76 2.52 3.80 

Folk dancing 3.64 2.54 4.15 3.10 2.86 3.45 

Games and sports 3.57 3.22 3.54 3.50 3.41 3.52 

Physical training 3.41 3.02 3.74 3.44 3.25 3.60 

Agriculture 3.30 3.20 3.95 4,00 3.28 3.96 

Cooking 3.06 3.63 3.06 3.63 

Manual training 3.49 3.60 3.49 3.60 

Sewing 3.11 3.41 3.11 3.41 

Shop work 3.57 3.50 3.57 3.50 

General Science 2.51 2.05 3.66 3.89 2.34 3.78 

Geography 2.17 2.36 3.60 3.73 2.26 3.67 

Nature Study 2.49 2.72 3.61 3.71 2.57 3.66 

Physiology and Hygiene 2.68 2.78 3.04 3.55 2.72 3.28 

General Average* 2.78 2.54 3.80 3.58 2.68 3.69 

Application 2.19 2.37 3.56 3.06 2.27 3.31 

Deportment 2.53 2.13 3.42 2.70 2.36 3.08 

*The "general average" is the mean of all the ratings given to the children of 
each group. 



EDUCATIONAL HISTORY 261 

It is clear that gifted children do work of superior quality 
in those subjects which require abstract thought, but they 
do only slightly better than average in those subjects which 
depend primarily upon manual dexterity or special talent. 





TABLE 93 






ORDER OF SCHOOL SUBJECTS WITH RESPECT TO DIFFERENCE 


IN QUALITY 


OF WORK OF 


GIFTED AND CONTROL 


GROUPS 






All 


All 




Subject 


Gifted 


Control 


Difference 


1. Debating or Speaking 


2.04 


3.88 


1.84 


2. U. S. History 


2.11 


3.92 


1.81 


3. Composition 


2.25 


3.93 


1.68 


4. Literature 


1.90 


3.56 


1.66 


5. Ancient History 


2.04 


3.60 


1.56 


6. Grammar 


2.25 


3.76 


1.51 


7. General Science 


2.35 


3.78 


1.43 


8. Geography 


2.26 


3.67 


1.41 


9. Civics or Citizenship 


2.12 


3.49 


1.37 


10. Reading 


1.92 


3.26 


1.34 


11. Arithmetic 


2.52 


3.80 


1.28 


12. Spelling 


2.13 


3.39 


1.26 


13. Dramatics 


2.40 


3.64 


1.24 


14. Nature Study 


2.57 


3.66 


1.09 


15. Agriculture 


3.28 


3.96 


.69 


16. Singing 


3.24 


3.89 


.65 


17. Folk Dancing 


2.86 


3.45 


.59 


18. Cooking 


3.06 


3.63 


.57 


19. Physiology or Hygiene 


2.72 


3.28 


.56 


20. Instrumental Music 


2.89 


3.38 


.49 


21. Physical Training 


3.25 


3.60 


.35 


22. Sewing 


3.11 


3.41 


.30 


23. Drawing 


3.62 


3.87 


.25 


24. Modeling 


3.48 


3.64 


.16 


25. Penmanship 


3.79 


3.92 


.13 


26. Games and Sports 


3.41 


3.52 


.11 


27. Manual Training 


3.49 


3.60 


.11 


28. Painting 


3.71 


3.80 


.09 


29. Shop Work 


3.57 


3.50 


-.07 



Even in these subjects, however, they excel the control group 
to some extent. It is a little surprising that reading and 
arithmetic do not rank somewhat higher than they do in 
amount of difference shown between gifted and control 
groups. 



262 TRAITS OF GIFTED CHILDREN 

In the above comparisons it should be remembered that 
the rating is made in comparison with other children of the 
same school grade, which ordinarily means that the gifted 
child is being compared with children one or two years older. 
This should more than counterbalance any possible "halo 
effect" entering into the ratings. 

Correlations (Pearson r's) between the various groups 
were computed for mean ratings earned in the different 
subjects. In working these correlations when the sexes were 
compared, cooking was paired with shop work, and manual 
training with sewing. The results are as follows: 

Gifted boys vs. gifted girls .891 .025 

" " control girls -.258 .111 

" " control boys .305 .116 

Control boys vs. control girls .204 .124 

" " gifted girls .279 .120 

Gifted girls vs. control girls .073 .165 

Total gifted vs. total control .207 .124 

The striking fact is the high correlation between gifted 
boys and gifted girls (.891), as compared with the low corre- 
lation for all other pairs. Giftedness is evidently far more 
potent than sex in determining relative success in the differ- 
ent school subjects. 

Table 94 shows for each subject the per cent of ratings 
which were high (1 or 2) and the per cent which were low 
(6 or 7). As would be expected, the differences between 
gifted and control are more obvious in the extremes than in 
the means. 

Additional information on subjects in which weakness 
was shown was obtained from the question given below: 

If child is especially weak in any subjects, give reason, 
if any reason is known to you. School Blank: II, 8. 

The following per cents were reported weak in one or 
more subjects: Gifted boys, 22 per cent; gifted girls, 15 per 
cent; control boys, 30 per cent; control girls, 18 per cent The 
gifted are weakest in the subjects which 'require manual 
dexterity. Writing, art, and hand work account for 68 per 
cent of the weaknesses reported for the gifted, as against 16 
per cent for the control. The control children are most often 
weak in the subjects requiring abstract thought. Arithmetic, 
reading, English, and history account for 61 per cent of the 



EDUCATIONAL HISTORY 



263 



TABLE 94 

PER CENT OF HIGH AND Low RATINGS ON SCHOOL WORK IN THE 
VARIOUS SUBJECTS 

(Per cents of ratings given on each subject, not per cents of children) 



All gifted 
N 1 or 2 6 or 7 



All control 
N 1 or 2 6 or 7 



1. Debating or Speaking 

2. United States History 

3. Composition 

4. Literature 

5. Ancient History 

6. Grammar 

7. General Science 

8. Geography 

9. Civics or Citizenship 

10. Reading 

11. Arithmetic 

12. Spelling 
[3. Dramatics 
14. Nature Study 
[5. Agriculture 

16. Singing 

L7. Folk Dancing 

18. Cooking 

[9. Physiology and Hygiene 

50. Instrumental Music 

51. Physical Training 
12. Sewing 

53. Drawing 

54. Modeling 

25. Penmanship 

26. Games and Sports 
11. Manual Training 

28. Painting 

29. Shop Work 

weaknesses reported for the control, as against 17 per cent 
For the gifted. Girls more often than boys were weak in 
arithmetic and history, and boys more often than girls in art 
and reading. This agrees fairly well with the distribution of 
low ratings in Table 94. The number of weaknesses re- 
ported was not large enough to give reliable rank orders of 
the subjects for the sex and intelligence groups. 1 

Additional data in Chapter XII on strength and weakness of gifted 
children in the different school subjects. 



215 


69.3 





114 


11.4 


5.3 


272 


64.3 


0.4 


192 


12.5 


6.3 


441 


60.8 


0.2 


369 


11.7 


5.4 


301 


71.8 





187 


17.1 


3.2 


102 


66.7 





63 


19.0 


12.7 


319 


60.8 


0.3 


216 


9.7 


2.8 


63 


54.0 





28 


3.6 





337 


57.6 





264 


9.8 


3.8 


107 


65.4 





57 


14.0 


1.8 


501 


74.3 


0.4 


472 


20.8 


4.4 


479 


51.4 


0.6 


446 


11.7 


4.7 


488 


64.1 


0.4 


489 


20.2 


3.7 


123 


57.7 


1.6 


80 


12.5 


2.5 


207 


42.5 





225 


8.4 


1.8 


35 


31.4 


2.9 


55 


3.6 


1.8 


397 


27.2 


6.3 


420 


12.6 


6.0 


75 


40.0 


1.3 


68 


27.9 


4.4 


88 


29.5 





77 


18.2 





39 


46.2 





36 


8.3 





128 


40.6 


0.8 


106 


25.4 


3.8 


389 


25.4 


3.1 


361 


14.4 


2.8 


107 


27.1 





115 


26.1 


1.7 


452 


18.4 


5.8 


447 


12.8 


6.0 


41 


17.1 


4.9 


39 


25.6 


2.6 


488 


18.0 


8.0 


437 


10.8 


9.2 


225 


23.6 


4.9 


257 


14.4 


1.6 


166 


19.9 


5.4 


122 


15.6 


2.5 


234 


15.8 


6.4 


189 


12.7 


4.8 


30 


26.7 


3.3 


30 


16.7 


3.3 



264 TRAITS OF GIFTED CHILDREN 

The reasons given by teachers for weakness in the various 
subjects were tabulated separately for the gifted and control 
groups by sex and subject, as follows: 

Gifted Control 

Coordination poor (manual subjects) 14 4 

Health or other physical handicap 13 15 

Lack of application 11 15 

Lack of interest 7 8 

Nervousness 6 4 

Inadequate previous training 5 9 

Carelessness 4 4 

Skipped grades 4 2 

Youthfulness (in games, etc.) 4 1 

Lack of special talent (as in music) 3 1 

Dislike of subject 2 2 

Speech defect 1 3 

Inaccuracy 1 2 

Lack of concentration or poor memory 3 

Language handicap 6 

Miscellaneous 2 

Is child's mental ability very even, ordinarily even, 
rather uneven, very uneven? (Underline.) Home Blank: 
II, 20. School Blank: IV, 2. 

TABLE 95 
EVENNESS OF ABILITY 

Very Ordinarily Rather Very Number 
even even uneven uneven reported 

Gifted boys, Home Blank 73.6% 24.1% 1.6% 0.6% 311 

Gifted girls, " " 78.9% 19.9% 0.8% 0.4% 261 

Total, " " 76.0% 22.7% 1.2% 0.5% 572 

Gifted boys, School Blank 71.7% 21.4% 3.3% 3.6% 304 

Gifted girls, " " 72.0% 23.2% 4.1% 0.8% 246 

Total, " 71.8% 22.0% 3.6% 2.0% 550 

Control boys, School Blank 22.3% 66.0% 10.5% 1.2% 256 

Control girls, " " 35.0% 56.0% 8.2% 0.8% 243 

Total, " " 28.5% 61.1% 9.4% 1.0% 499 

Outstanding facts in the above table are the following: 

(1) A high percentage of gifted, as compared with the 
control, are rated as "very even." The ratio is almost three 
to one. 

(2) The school rates, 5.6 per cent of the gifted, as com- 
pared with 10.4 per cent of the control, as either "rather un- 



EDUCATIONAL HISTORY 



265 



even" or "very uneven." This is in the ratio approximately 
of one to two. 

(3) Girls are rated as "very even" more often than boys, 
and somewhat less often as very uneven. 

(4) The ratings of school and home on the gifted agree 
very closely with regard to numbers reported in each 
category. 

The above data have only suggestive value, as many dis- 
crepancies and inconsistencies were found. In Chapter XII it 
is shown on the basis of test results that gifted children have 
about the same tendency to unevenness as unselected chil- 
dren of corresponding school grade or of corresponding 
mental age. 

Question IV, 3 in the School Blank, and II, 21 in the Home 
Blank, asking in what respects ability was especially strong 
or especially weak, was so frequently misunderstood that 
the results could not be used. 

Could child do school work of average quality if pro- 
moted to a higher grade now and given a certain amount of 
coaching on the subject matter skipped? . . . If so, in what 
grade as highest? . . . School Blank: II, 5. The results are 
given in Table 96. 



TABLE 96 


EXTRA PROMOTION MERITED 


Group 


(Teachers* judgments) 

Per cent that could be advanced 
No. Re- various numbers of half grades 
ported "Yes" 1234 


5 


Average 
amount of 
deserved 
promotion 


Gifted Boys 
Gifted Girls 
Control Boys 
Control Girls 
All Gifted 
All Control 


269 
217 
239 
228 
486 
467 


84% 
79.3% 
36% 
44.3% 
81.9% 
40.5% 


38.3% 
36.4% 
28.4% 
32.0% 
37.4% 
30.2% 


30.9% 
23% 
6.7% 
9.2% 
27.4% 
7.9% 


4.8% 
5.5% 
0.4% 

5.2% 
0.2% 


1.9% 
1.8% 


1.9% 




0.9% 


0.4% 



1.3 Half grades 
1.2 " 
0.4 " 
0.5 " 
1.3 " 
0.5 " 



The most striking facts brought out are (1) that in the 
judgment of the teachers, 82 per cent of the gifted group are 
prepared to do the work of a higher grade, as compared 
with 40 per cent of the control group; (2) that the extra 
grade advancement which these 82 per cent of gifted chil- 
dren are said to merit, is 1.4 half grades, or about three- 



266 TRAITS OF GIFTED CHILDREN 

quarters of a grade. The corresponding figure for the 40 per 
cent of control children who are said to merit extra promo- 
tion is .5 of a half-grade, or one quarter of a grade. 

If over-promotion had also been taken account of, the 
difference between the two groups would have been even 
more pronounced. The teachers were not asked to report on 
those children who, in their opinion, were already advanced 
beyond their ability. Since of the gifted group 82 per cent 
are reported able to do work of a higher grade, while of the 
control group only slightly over 40 per cent are so reported, 
it follows that (to take an extreme illustration) even if no 
children were correctly graded, only 18 per cent of the gifted 
group could be graded too high, while of the control group 
nearly 60 per cent might be. 

EDUCATIONAL HISTORY 

The data on educational history have come chiefly from 
the Home Information Blank, which devoted two pages to 
the subject. Such reports cannot, of course, be taken as 
entirely accurate for individual children, but for the group 
as a whole they are believed to be fairly dependable. There is 
every evidence that as a rule the blanks were filled out with 
painstaking conscientiousness. In some of the blanks a few 
of the questions were not answered, but the per cent of total 
for whom a question is answered is usually 85 per cent and 
95 per cent. As the Home Blank was not filled out for a 
control group, the only comparative data available are such 
as may be found in educational literature. 

It will be understood that the figures given in the follow- 
ing pages relate to the 643 children of the main experimental 
group. 

Did child attend kindergarten? . . . When and how 
long? Home Blank: II 9 1. Reports for 314 boys and 258 
girls. Answer is "yes" for 62.4 per cent of boys and 60.1 per 
cent of girls. Of 351 who had attended kindergarten, the 
length of attendance was stated for all except 14. The mean 
length of attendance was 5.9 months for boys and 5.5 months 
for girls. This is slightly more than half a school year. 

Age of entering school (above kindergarten) . . . Grade 
first entered . . . Home Blank: II, 2. The first part of this 
item was answered for 541 cases, the second part for 533. 



EDUCATIONAL HISTORY 267 

As shown in Table 97, the mean age of entrance was about 
6% years. 

TABLE 97 
AGE OF ENTERING SCHOOL AND GRADE FIRST ENTERED* 

Age of entering school, 

above kindergarten Grade first entered 

Age No. Cases Grade Cases 

1 L 421 

1 H 57 

2 L 26 

2 H 12 

3 L 5 

3 H 5 

4 L 5 

Total 541 JL J 



4-0 to 4-11 


13 


5-0 to 5-11 


154 


6-0 to 6-11 


328 


7-0 to 7-11 


39 


8-0 to 8-11 


6 


9-0 to 9-11 





10-0 to 10-11 


1 



Total 533 

*For age, median is 6 yrs., 3.79 mos.; mean is 6 yrs., 3.24 mos. ; S. D. is 7.28 
mos. For grade, median is .63 and mean is .44 of the distance through the low 
first; S. D. is 1.12 half grades. 

Name and location of schools child has attended, in- 
cluding kindergarten. Home Blank : II, 5. The ranges and 
means of the number of schools attended by children of the 
different ages, sexes combined, are as follows : 

Age 5 6 7 8 9 10 11 12 13 

Range 1-3 1-4 1-4 1-4 1-6 1-7 1-11 1-7 1-6 
Mean 1.53 1.46 1.70 2.07 2.43 2.67 3.14 3.29 3.54 

The one child who had attended eleven different schools 
(in six different cities and three different states) was, at 
the age of eleven years three months, finishing the work of 
the low seventh grade with decidedly better than average 
standing in all major subjects. His poorest work according to 
the teacher's report is in writing and physical training, in 
each of which he is graded as "low average." He has attended 
school above kindergarten a total of five years and has 
skipped three half grades. He did not learn to read before 
starting to school, and has had no home instruction in any 
of the regular school subjects, although for the past five 
years he has had private lessons in French in addition to his 
school work. The teacher reports that he might safely be 
advanced another half year in school. Such cases (there 



268 TRAITS OF GIFTED CHILDREN 

were several others almost as striking as this one) suggest 
that change of schools, which is so often reported in educa- 
tional literature as one of the main causes of retardation, 
may be a very minor factor compared with native ability. 
Gifted children are doubtless able to adapt themselves more 
readily than normal children to a new environment, and it is 
even possible that the benefits which the children of our 
group have received from such changes, in the way of broad- 
ened experience and interest, outweigh the usual disadvan- 
tages. This might not be true of normal children. 

The data on name and location of schools attended were 
secured primarily for use in the special study of individual 
cases and need not be summarized here. 

Average number of school days missed in a year (rough 
estimate). . . Any long absences? . . . How long? . . . 
Reasons for irregularity. . . . Home Blank: II, 6. The re- 
sults are as follows for the average estimated number of 
days missed : 

Boys Girls Total 

No. for whom answered 238 201 439 

Total range of days missed 0-80 0-70 0-80 

Median 9.65 9.05 9.35 

Mean 13.45 11.10 12.39 

S.D. 13.75 10.05 12.25 

The question regarding long absences was answered for 
521 cases. For 55 per cent of these it is reported that no 
long absences have occurred. For 45 per cent, absences 
varying in length from two weeks to one and a half years 
are reported. Of long absences reported, the length of 22 
per cent is not stated. In 57 per cent of the cases the ab- 
sence was less than three months; in 15 per cent from three 
to six months; and in 6 per cent over six months. A slightly 
higher percentage of long absences is reported for boys than 
for girls. 

Reasons for irregularity are given in 223 cases, as fol- 
lows: 

Reason No. Cases Per cent 

Illness (personal) 178 76.4 

Illness in family 7 3.0 

Travelling 30 12.9 

Change of schools 6 2.6 

Ahead of class 3 1.3 

Miscellaneous 9 3.8 



EDUCATIONAL HISTORY 269 

Has attendance been very regular, fairly regular, rather 
irregular, very irregular? (Underline.) School Blank: IV, 
7. The data from this question are available for the control 
group as well as for the gifted. The results showed no sig- 
nificant differences between gifted and control or between 
boys and girls. 

Cases Very Fairly Rather Very 

reported regular regular irreg. irreg. 

Control Boys 245 72% 22% 4% 1% 

Girls 263 74% 21% 3% 2% 

Gifted Boys 312 79% 14% 5% 2% 

" Girls 263 80% 15% 3% 2% 

Liking for school: very strong, fairly strong, slight liking, 
positive dislike. (Underline.) If school has been disliked at 
any time, why? . . . . Home Blank: II, 7. 



No. 
reported 


Very 
strong 


Fairly 
strong 


Slight 
liking 


Positive 
dislike 


310 
254 
564 


54.5% 
70.1% 
61.5% 


39.4% 
26.7% 
33.7% 


5.2% 

2.4% 
3.9% 


1.0% 
0.8% 
0.9% 



Gifted Boys 
" Girls 
All Gifted 

Only 5 of the 564 children in the gifted group are re- 
ported as positively disliking school. Very strong liking for 
school is more common among girls than among boys. The 
reasons for dislike, however, disclosed no sex difference. 
The following reasons are reported : 

No. of times 
Reason reported 

Disliked teacher 39 

Daily monotony 8 

Change of school 3 

Work too elementary 6 

Work too difficult 2 

Timidity 2 
Miscellaneous, including distance from school, 

overcrowded conditions, "prefers to play," 

"dislikes compulsion", etc. 18 

These seem to be entirely normal reasons. 

Describe child's attitude toward school. School Blank: 
II, 7. In order to treat the heterogeneous mass of descriptive 
matter included under answers to this question, it was nec- 
essary to condense and categorize the replies. In deciding 
upon categories the principle followed was, first, to adhere 



270 TRAITS OF GIFTED CHILDREN 

to the original wording wherever possible, and, secondly, 
where answers were too long to permit this, to select the 
key word of the sentence (an adjective as a rule) and cate- 
gorize under this. The responses to this question are prob- 
ably quite as much subjective as objective, and indicate the 
teacher's attitude toward the children as much as the chil- 
dren's attitude toward school and the teacher. They are 
none the less interesting on this account. It will be noted 
(Table 98) that the sex difference is somewhat greater than 
the intellectual difference. Other points of interest are : 

(1) The excess of gifted over control described as ex- 
cellent, enthusiastic, eager to excel, or inattentive; 

(2) The excess of control over gifted described as lack- 
ing interest or showing unsatisfactory attitude; 

(3) That undesirable attitudes are twice as prevalent in 
the control group as in the gifted group; 

(4) That no gifted girls are described as inattentive, but 
that five gifted boys are so described. 

TABLE 98 
ATTITUDE TOWARD SCHOOL, AS DESCRIBED BY TEACHERS 

Gifted Group Control Group 

Boys Girls Total Boys Girls Total 

Desirable attitudes 



Conscientious 


12 


4 


16 


5 


6 


11 


Interested 


130 


104 


234 


92 


115 


207 


Excellent 


25 


34 


59 


13 


17 


30 


Cooperative 


6 


2 


8 


5 


3 


S 


Enthusiastic 


5 


14 


19 


4 


3 


7 


Eager to excel 


6 


9 


15 


3 


5 


8 


Studious 


11 


5 


16 


3 


7 


10 


Happy 


3 


9 


12 


8 


10 


18 


Excellent work 














with normal 














play interest 


2 


1 


3 











Good 


30- 


20 


50 


37 


31 


68 


Normal 


17 


16 


33 


20 


10 


30 


Fairly interested 














3 


3 



Total with de- 
sirable attitudes 247 218 465 190 210 400 

Number in 
each group 271 227 498 236 227 463 

Per cent showing 
desirable attitude 91.1% 96.0% 93.4% 80.5% 92.5% 86.4% 



EDUCATIONAL HISTORY 271 

TABLE 98 Concluded 

Gifted Group Control Group 

Boys Girls Total Boys Girls Total 

Undesirable attitudes 
Lacks interest 10 6 16 34 15 49 



Reticent 


3 





3 





1 


1 




Inattentive 


5 





5 













Mischievous 


1 





1 













Conceited and lazy 


2 


1 


3 


3 





3 




Attitude poor 


1 





1 


8 





8 




Spoiled 


1 





1 













"Self-satisfied" 


1 


1 


2 













Precocious 





1 


1 













Sullen 











1 





1 




Dislikes school 














1 


1 




Total undesira- 
















ble attitude 


24 


9 


33 


46 


17 


63 




Number in 
















each group 271 


227 


498 


236 


227 


463 




Per cent showing 
















undesirable 
















attitude 


8.9% 


4.0% 


6.6% 


19.5% 


7.5% 


13 


.6% 



Eelative proportions showing undesirable attitudes: 

Ratios 

Gifted boys, control boys 1 to 2.2 

Gifted girls, control girls 1 to 1.9 

Gifted group, control group 1 to 2.1 

Gifted girls, gifted boys 1 to 2.2 

Control girls, control boys 1 to 2.6 

Did child learn to read before starting to school? . . . At 
what age? . . . How (amount and kind of help)? . . . 
Home Blank: II, 12. Of 300 boys for whom report was 
made, 44.3 per cent learned to read before starting to 
school; of 252 girls, 46.4 per cent. The more or less exact 
age of learning to read is reported for 246 cases. (Table 99.) 

Probably the way in which the question is worded ac- 
counts for the fact that less than half the parents gave the 
age of learning to read. For the most part those who re- 
ported that the child did not learn to read before starting to 
school, did not answer the second part of the question. Since 
so little precise information is obtainable as to how long a 
period elapses after entering school before the average 
child may be said to have learned to read, and still less in- 
formation as to what would be the corresponding time for 



272 TRAITS OF GIFTED CHILDREN 

gifted children, any statistical work based upon such in- 
formation as we have would be extremely inaccurate. We 
can say, however, that at least 113 (20.5 per cent) of the 552 
children learned to read before the age of 5; 34 (6.1 per 
cent) before 4; and 9 (1.6 per cent) before 3. 

TABLE 99 
AGE OF LEARNING TO READ 



Age 


Boys 


Girls 


Tot 


1-0 to 1-5 


1 





1 


1-6 to 1-11 


1 





1 


2-0 to 2-5 


2 


2 


4 


2-6 to 2-11 


1 


2 


3 


3-0 to 3-5 


8 


6 


14 


3-6 to 3-11 


8 


3 


11 


4_0 to 4-5 


25 


28 


53 


4-6 to 4-11 


15 


11 


26 


5_0 to 5-5 


53 


40 


93 


5-6 to 5-11 


10 


11 


21 


6-0 to 6-5 


7 


9 


16 


6-6 to 6-11 


1 





1 


7-0 to 7-5 





1 


1 


7-6 to 7-11 





1 


1 


Total 


132 


114 


246 



On the last part of the question (amount and kind of 
help) answers were received for 197 children. (This ques- 
tion applies only to the children who are reported as having 
learned to read before entering school.) In a large majority 
of cases, only incidental or casual assistance is reported. In 
5 cases the child had a private tutor, in no case for more 
than a half hour daily. In 17 cases the child is said to have 
"taught himself" without the knowledge of any member of 
the family, until it was suddenly discovered that the child 
could read. "By playing with books, pictures, or news- 
papers" is a rather frequent answer. Very informal assis- 
tance from some member of the household, usually the 
mother, is reported for 129 cases. In many cases it is speci- 
fied that this help was given only in response to urgent so- 
licitations on the part of the child. 

Underline each half grade skipped: IB, 1A, W, 2A, 3B, 
3A, bB, bA, 5B, 5A, 6B, 6A, 7B, 7 A, 8B, 8A. Home Blank: 
11,3. 



EDUCATIONAL HISTORY 273 

Underline each half grade repeated: IB, 1A, 2B, 2A, 3B, 
3A, 4B, M, 5B, 5 A, 6B, 6A, 7B, 7 A, 8B, 8A. Home Blank: 
11,4. 

Boys Girls Total 

Number of cases reported 301 249 550 

Per cent who have skipped 82.7 85.9 84.9 

Per cent who have repeated 3.7 4.4 4.0 

Mean net gain in half grades 1.91 2.17 2.03 

Median" " " " " 1.10 1.38 1.24 

The mean net acceleration is based upon the number of 
half grades skipped minus the number repeated. From the 
following figures it will be seen that the mean net accelera- 
tion increases very little after the age of 7 years. 

All 
Age 6 7 8 9 10 11 12 13 ages 

Mean net accel- 
eration, Boys 0.67 1.84 1.93 1.84 2.05 2.42 1.92 1.08 1.91 

Mean net accel- 
eration, Girls 0.69 1.62 1.86 2.53 2.40 2,56 2.72 2.32 2.17 

Maximum net 
acceleration, 
Boys 387762548 

Maximum net 
acceleration, 
Girls 345856 10 5 10 

The important fact in the above figures is that the aver- 
age net gain by skipping amounts to only about one school 
grade, and that nearly all of this gain is made before the age 
of eight years. After eight years the net gain is more likely 
to increase with girls than with boys. The above figures 
agree very well with the data presented in the first section 
of this chapter. 

Have you encouraged the child to forge ahead in school, 
allowed him to go his own pace, or held him back? .... 
Why? .... Home Blank: II, 9. This question was answered 
by 550 parents, of whom 396 gave reasons for their attitude. 
The data are as follows : 

Have encouraged child to forge ahead 106, or 19.3% 

Have allowed child to go his own pace 392, or 71.3% 

Have held child back 52, or 9.4% 



274 TRAITS OF GIFTED CHILDREN 

Reasons given for encouraging rapid progress: 

Realize necessity of education 16 

Natural parental interest 15 

Dislike laggards 9 

Want him to like school 7 

To keep him busy mentally 4 

To make up for absences 1 

Work too elementary to stimulate interest 1 

To encourage ambition 1 

Reasons given for allowing own pace: 

Forcing unnecessary 149 

Do not believe in forcing 52 

Believe in normal development 1 

Youth 36 

Health 14 

Afraid to interfere 27 

Mother busy, no time to supervise 2 

Parents do not speak English 1 

Want him to like school 5 

To develop individuality 3 
Poor instruction made rapid progress 

inadvisable 1 

Social problems (no explanation) 1 

Reasons given for holding child back: 

Youth 12 

Afraid of overstrain 15 

Health 14 

< Advice of school authorities 5 

Child stutters 1 

To give more time for recreation 1 

To work in manual training 1 

To keep in school near home 1 

There may possibly be some significance in the fact that 
only 51 per cent of the first group give a reason for the 
course which they report they have followed, as compared 
with 75 per cent of the second group and 96 per cent of the 
third group. It is clear, however, that the large majority of 
our subjects have not been subjected to "hothouse" methods 
and that this factor does not account for their school accel- 
eration or for their high score in the intelligence test. 

Average number of HOURS A WEEK devoted to HOME 
STUDY of school lessons during the last year. Home Blank: 
II, 8. Reports were received for 271 boys and 221 girls. 
Tables 100 and 101 give the results separately by age and 
sex. It is interesting to note the marked sex difference and 



EDUCATIONAL HISTORY 275 

the large number of cases in the zero column. It would ap- 
pear that home study has not played a very large part in the 
almost universally good school records of these children. 
The mean for the entire group is less than two hours a week. 
The variability, however, is unusually large. In general the 
girls do more than twice as much home study as the boys. 

TABLE 100 

HOURS A WEEK IN HOME STUDY OF SCHOOL LESSONS 
(GIFTED BOYS) 



Hours 


No. 




Age 





1 


2 


3 


4 


5 


6 


7 8 9 10 11 12 13 14 15 


cases 


Mean 


5 


5 
















5 


0. 


6 


11 


1 


1 












13 


0.15 


7 


13 


3 


2 




1 




1 




20 


0.68 


8 


24 


4 


3 


1 


1 








33 


0.38 


9 


27 


8 


7 


5 


2 






1 1 


51 


0.94 


10 


21 


11 


5 


6 


3 


5 


3 


1 1 


56 


1.76 


11 


13 


10 


10 


10 


1 


3 


4 


1 1 


53 


1.89 


12 


8 


7 


2 


5 




5 


1 


2 1 


31 


2.21 


13 




1 


2 


1 


1 


1 


2 


1 


9 


3.63 



Total 122 45 32 28 9 14 11 5 1 1 2 1 271 1.39 

Limits of groups Zero indicates no home study. 

1 Indicates any amount up to and including one hour. 
2 Indicates from one up to and including two hours, etc. 

TABLE 101 

HOURS A WEEK IN HOME STUDY OF SCHOOL LESSONS 
(GIFTED GIRLS) 

Hours No . 

Age 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 cases Mean 



5 
6 


9 






1 




1 















11 


0.64 


7 


16 




1 


1 


















18 


0.25 


8 


14 


5 


3 


2 


1 


1 


1 






2 


1 




30 


1.86 


9 


19 


6 


3 


5 


1 


1 






1 


1 






37 


1.08 


10 


11 


3 


10 


2 


2 


5 


2 


2 






2 


1 


40 


2.81 


11 


9 


8 


5 


7 


2 


2 


3 


3 








1 


40 


2.36 


12 


7 


1 


7 


3 


1 


3 


2 




2 


3 1 


1 




31 


3.61 


13 


1 


1 


2 


1 


3 






1 


1 


1 


1 


2 


14 


5.83 


Total 


86 


24 


31 


22 


10 


13 


8 


6 


3 1 


7 1 


5 


1 3 


221 


2.32 



Private tutoring (out of school) Home Blank: II, 11. 
After the question were blank spaces for recording informa- 
tion on tutoring in music, drawing, painting, dancing, lan- 
guage, and "other subjects." The information called for in- 
cluded age at which the instruction was taken, hours a week 



276 TRAITS OF GIFTED CHILDREN 

(including practice), and a rating of the ability shown as 
very superior, superior, average, inferior, or very inferior. 
Of the 597 in the main group for whom Home Blanks 
were received, private tutoring was reported for 338, or 56 
per cent. For the girls the proportion was 72 per cent; for 
the boys, 44.4 per cent. Those who had private tutoring de- 
voted to it an average of 6.5 hours a week, including prac- 
tice. Failure to reply to this question was taken to indicate 
that no private tutoring had been received, as this interpre- 
tation seemed valid in most cases. The above figures, how- 
ever, may be somewhat too low. Even so, it would take a 
considerable stretch of the imagination to ascribe the super- 
ior intellectual development of these children to the results 
of private tutoring. Following is a detailed summary of the 

reports. 

TABLE 102 

AMOUNT AND RESULTS OF PRIVATE TUTORING 

Median Ability shown 

Percent hours Very Very Not 

taking -weekly sup. Sup. Av. Infer, infer. stated 

Music 



Boys 34% 


5.4 


11% 


34% 


47% 


2% 





6% 


Girls 59% 


5.9 


13% 


38% 


40% 


2.5% 


1% 


6% 


Drawing and 
















painting 
















Boys 3% 


5 


20% 


60% 


40% 











Girls 3% 


2 





63% 


25% 








12% 


Dancing 
















Boys 10.6% 


1.6 


9% 


14% 


51% 


12% 





14% 


Girls 32.7% 


1.8 


15% 


22% 


52% 


2% 


1% 


8% 



Language 

Boys 7.3% 3 12% 25% 42% 21% 

Girls 10.4% 2.7 14% 32% 39% 14% 

"Other subjects" 

Boys 6.4% 3 
Girls 8.9% 2.7 

In the following spaces, put down a rough guess at the 
average number of hours a week of instruction child has 
received from members of the household at the ages indi- 
cated. Home Blank: II, 10. After the question were spaces 
for such records for ages 2-3, 4-5, 6-7. Kinds of instruction 
mentioned after the question were "telling stories or read- 



EDUCATIONAL HISTORY 277 

ing to child," "teaching to read and write," "number work," 
and "nature study." 

This question has been answered so incompletely as to 
make statistics misleading. Parents in many instances re- 
port that it is impossible for them to make even an approxi- 
mate estimate, as the time spent in this way was so inciden- 
tal and varied so much from day to day. In a large percent- 
age of cases the question has been left unanswered. If the 
mean time were worked out for those answering the ques- 
tion, omitting those cases for whom it was not answered, the 
figures would be too high; if the cases for whom the ques- 
tion was not answered were counted as zero, the mean 
would be too low. 

In the 595 blanks which have been returned, the largest 
number of replies to the first part of the question (telling 
stories or reading to child) at any of the three age periods 
for which the report is asked, is 380. To the second part 
(teaching child to read or write), the largest number of re- 
plies is 151. To the third part (number work with child), 
the largest number of replies is 146. To the fourth part 
(nature study work with child), the largest number of re- 
plies is 98. These figures seem to indicate that in the ma- 
jority of cases failure to reply should be interpreted as 
indicating that no time was spent in this way, since the 
above order corresponds roughly at least to what one would 
expect in the way of negative responses. It is probable, how- 
ever, that such is not always the case and that to interpret 
the entire group of failures to reply in this way would be to 
introduce a considerable element of error. 

Has child shown very superior ability with respect to: 
(a) General intelligence? .... Age when first noted? . . . 
How shown? .... (Similarly for music, arithmetic or 
mathematics, science or nature study, mechanical ingenu- 
ity, drawing or painting, dramatics, dexterity in hand work.) 
Home Blank: II, 19. Table 103 gives the results for the 
614 children for whom Home Blanks were received (334 
boys and 280 girls) . The "yes" and "no" columns giving per 
cents of total are perhaps the most significant. Among the 
outstanding facts are the following : 

1. Girls more often than boys have shown superior abil- 
ity in dramatics, music, and dexterity in hand work. 



278 



TRAITS OF GIFTED CHILDREN 



2. Boys more often than girls have shown superior abil- 
ity in mechanical ingenuity and (to a slight extent) in arith- 
metic or mathematics. 

3. Apart from general intelligence, superior ability is 
most often shown in arithmetic or mathematics in the case 
of both boys and girls. 

4. In 8 per cent of cases, both boys and girls, the parents 
definitely state that they had not observed any indications 
of superiority in general intelligence. This rather surpris- 
ing fact is probably accounted for by the high standard of 
intelligence which prevails in the home. 



TABLE 103 

PROPORTION OF GIFTED REPORTED BY PARENTS AS SHOWING VARIOUS 
KINDS OF SUPERIOR ABILITY 



Superior 
Ability 

General Intel. 

Music 

Arith. or Math. 

Sci. or Nat. Std. 

Dramatics 

Drawing & Paint. 

Dexterity, Hand 

Work 
Mech. Ingenuity 



Not 


"No" 


"Yes" 


"Yes" 


Stated 


% of Total 


% of Total 


% of answered 


Boys 


Girls 


Boys 


Girls 


Boys 


Girls 


Boys 


Girls 


8% 


7% 


8% 


8% 


84% 


85% 


92% 


91% 


24% 


27% 


45% 


31% 


31% 


42% 


41% 


57% 


22% 


27% 


27% 


28% 


52% 


45% 


66% 


62% 


43% 


43% 


28% 


28% 


29% 


29% 


51% 


51% 


41% 


41% 


41% 


17% 


18% 


42% 


31% 


71% 


38% 


45% 


35% 


24% 


27% 


31% 


43% 


57% 



46% 42% 33% 26% 21% 31% 39% 54% 
35% 50% 34% 43% 31% 7% 48% 15% 



The reports regarding age at which superior abilities 
were first shown are summarized in Table 104. 

1. The sex differences are here relatively small, but the 
lines in which superior ability is noted somewhat earlier 
with the boys than with the girls include music, arithmetic 
or mathematics, drawing or painting, and dexterity in hand 
work. The girls tend to show superior ability earlier in gen- 
eral intelligence, science or nature study, .and mechanical 
ingenuity. 

2. Considering both sexes, the order shown by the vari- 
ous abilities for precocity is, first, general intelligence 
(mean about 3.5 years); second, music (mean about 4.9 
years) ; third, all the other abilities (means from about 5.75 
to about 6.25 years). 



EDUCATIONAL HISTORY 279 



X COCO CO rH CMCO CM OS C-lO D- CO CS - , 

W -^O >^CM O O* t-Ql-Q ^ ^ CM rH *** CO <P_ CO 

^CMCM CMCO CMrH CMCM CMCM 

S *^ CO OOOO OOCO UO"* OOXO CMO 

g f~:<=M COrH OO-* ^1^ O r*; "*<=2 ^? c . 

M COCO ^Ift 1OCO COVO CO^O VT5CT5 



W 
O 



Cfl 

w 



W rH rH 

H 

^ d CO CO CM OO O Oft 00 CM t- CMCM O 00 
O ^ rH T-H r-i CMCM rH rH rH 

tq e 



^ 



S ^ O5 o coo o o r- o i-t r-i oo> r* r-i 

P T-I CM i i T-I CO T- r-i 
CO 

OO CM CO \fi> O r- T-H UO CO O r- CR Oft OO O 



CM OO O OS OO C>- iO r- CO OO T-H O Oi COCO CO CM 
"^ 10 r-< i-H i-i 

I 

rH XO TH COCO rH rH -<*< rH CM CO CM 

00 CO CO 

o 

** 

rH CO Tt C5 CM rH CM rH 

d, "^ 7 

^Tt*!^ THOO r^-ift OO^ CMt^- COCft O5CO -^OO 
.2; -^i) O5O COO C ift COrH I>-CO -rHCi kOCO CM 
CM CM rH T-H rH 



PQO C3O CQO WO CQO P3 O O MO 



& . ^ " 

I il &i ij :. s ^| 5 

*% B 5 S S "3 .5 .2 "S "G ^ o 

si 1 Is |l || 1.1 s i fe 

5S I ^S %& II S2 a Sa 



280 



TRAITS OF GIFTED CHILDREN 



3. The sex differences in variability with respect to the 
age at which the various abilities are noted are too small to 
be very significant and are not uniform in direction. 

The descriptions which the parents give of the early in- 
dications of superiority are usually couched in such general 
terms as to be less enlightening than one could wish. How- 
ever, the tabulations are given for what they may be worth. 
The "miscellaneous" group includes all the indications 
which were not mentioned for as many as 5 per cent of the 
children in either sex for whom the special ability had been 
noted. 

TABLE 105 
EARLY INDICATIONS OF SUPERIOR ABILITY 

Boys Girls 

General Intelligence < 282 ) < 237 ) 

Grasps and understands new ideas quickly 50 40 

Desire for knowledge 31 31 

Retentive memory 21 21 

Intelligent conversation 20 15 

Rapid progress at school 16 16 

Keen general interests 22 9 

Range of general information 15 12 

Reasoning ability 13 13 

Early speech 11 14 

Asking intelligent questions 14 11 

Ability in accomplishing difficult things 14 11 

Keen observation 13 10 

Unusual vocabulary 8 12 

Originality 3 12 

Miscellaneous 4 6 

Positive report without explanation 27 14 

Boys Girls 

MUSIC (104) (117) 

Good ear for music 19 20 

Rapid progress in lessons 11 15 

Sang well at early age 14 11 

Carries tune well 8 9 

Persistent desire to learn music 8 9 

Sense of rhythm 8 8 

Great interest in music 10 5 

Unusual musical appreciation 4 10 

Learns music easily 6 5 

Has natural talent 7 4 

Good work in original composition 4 6 

Miscellaneous 6 

Positive report without explanation 5 9 



EDUCATIONAL HISTORY 
TABLE 105 Continued 



281 



Boys Girls 

Arithmetic or Mathematics (172) (126) 

Rapid progress at school 32 29 
Desire to learn advanced work 19 15 
Grasps numerical ideas quickly 19 8 
Learned to count and make number combina- 
tions at early age 19 7 
Is very quick with figures 7 14 
Does difficult work with no effort 9 11 
Unusual reasoning ability 8 12 
Could make change at early age 8 3 
Miscellaneous 25 14 
Positive report without explanation 26 13 



Science or Nature Study 

Great lover of nature 

Enjoys nature stories 

Studies plant life 

Collections of natural history specimens 

Great interest in flowers and trees 

Great interest in animals 

Many questions on scientific subjects 

Bird study 

Miscellaneous 

Positive report without explanation 



Boys 


Girls 


(98) 


(82) 


26 


26 


12 


3 


4 


9 


7 


6 


6 


7 


6 


4 


5 


4 


2 


4 


22 


9 


8 


10 



Mechanical Ingenuity 

Building with meccanno sets 

Mechanical construction 

Repairing things 

Designing toys 

Building boats, coasters, etc. 

Knows the makes of all cars 

Likes to work with tools 

Progress in manual training 

Mechanical drawing 

Reads about mechanics 

Studies ways of improving devices 

Interested in wireless 

Interested in anything mechanical 

By persistent questions about mechanics 

Miscellaneous 

Positive report without explanation 



Boys 

(105) 

16 
14 

6 
7 
7 
7 
6 
4 
5 
4 
4 
1 
3 

9 
12 



Girls 
(21) 

3 
2 
3 
1 



1 



2 

2 
3 
4 



282 



TRAITS OF GIFTED CHILDREN 



TABLE 105 Concluded 

Drawing or Painting 
Marked interest in drawing or painting 
Rapid progress in drawing or painting 
Good at cartooning 
Likes to copy pictures 
Illustrating 
Good original work 
Likes to sketch 
Natural talent 
Interested in designing 
Interested in mechanical drawing 
Miscellaneous 
Positive report without explanation 



Dramatics 

In school and Sunday School plays 

By impersonating characters 

By expressive reading 

By making up theatricals 

By reciting and telling stories 

In public speaking 

Marked interest in dramatics 

By writing plays 

Successful stage or screen acting 

Interpretive dancing 

Miscellaneous 

Positive report without explanation 



Boys 
(89) 

29 

10 

10 

4 

3 

4 

2 

1 

1 

5 

15 
5 

Boys 

(61) 

13 
10 

1 

5 

8 

5 

2 

1 

2 



6 

8 



Girls 

(87) 

33 
5 
4 
5 
6 
4 
6 
5 
4 

8 
7 

Girls 
(118) 

21 
11 
15 
10 

3 

6 

8 

8 

6 

7 
15 

8 



Dexterity in Hand Work 

Needlework or sewing 

All kinds of hand work 

Making dolls* clothes 

Manual training progress 

In school work 

Crocheting or knitting 

Marked interest in hand work 

Clay modeling 

Cutting 

Use of tools 

Making useful things 

Very neat and accurate 

Making toys 

Wood carving and modeling 

Miscellaneous 

Positive report without explanation 



Boys 
(70) 

1 

2 

9 
3 
1 
3 
3 
4 
6 
6 
2 
5 
4 
8 
13 



Girls 

(87) 

29 
14 
14 



5 

7 

3 

3 

2 





3 





4 

3 



EDUCATIONAL HISTORY 283 

What theories or ideas of child training have guided 
your educational efforts with the child? Please answer 
fully; for example, indicate what principles or rules have 
guided you in regard to each of the following: (A) answer- 
ing child's questions; (B) stimulating desire to learn; (C) 
other matters you have considered important. Home 
Blank: 11,16. 

This question is of interest only as showing general 
trends of thought. Replies have been rather loosely catego- 
rized and the categories arranged in order of frequency of 
occurrence. No further statistical treatment seemed worth 
while, as the chief value of the question lies in its import for 
the individual child rather than for the group as a whole. 

No age or sex divisions seemed advisable. Several re- 
plies have often been given to the same part of the question 
(especially Part C) ; therefore the total number of replies 
does not coincide with the number of children for whom 
the question was answered. Many parents failed to answer 
the question at all, many others answered Part A only. A 
few answers in cases where the meaning was not clear were 
thrown out. 

TABLE 106 

METHODS USED IN HOME TRAINING 
A. Answering child's questions: 

1. General replies such as "answer fully," "completely," 

"truthfully," "to the best of my ability," etc 354 

"Patiently" is a word which frequently occurs here, an 
eloquent expression of the frequency with which the 
questions occur. "To the verge of exhaustion," one parent 
reports. In seven cases, however, the parent adds that 

the child asks "very few questions." This is usually fol- 
lowed by the comment, "He prefers to find out things for 
himself, does not like to he told them." 

2. Answer if possible, otherwise look up matter with or for 

child or direct him to source of information 61 

As might be expected, this answer occurs far more fre- 
quently with the older children than with the younger. 

3. Try to suit answers to the level of his intelligence ... 55 

4. Answer questions which seem worth while. Discourage 
unnecessary questioning 15 

5. Answer in such a way as to encourage further study and 
investigation 14 

6. Answer completely unless matter is one concerning which 
child can and should find out for himself 13 



284 TRAITS OF GIFTED CHILDREN 

TABLE 106 Continued 

7. Do not answer questions on sex matters ...... 5 

8. Through concrete experience wherever possible. Let him 

see the process 5 

9. Answer if convenient, otherwise not 4 

10. Always try to compare with fa % cts already known ... 4 

B. Stimulating desire to learn: 

1. Stimulating not necessary 97 

One parent adds, "On the contrary the child himself is, 

for the entire family, the greatest possible 'stimulation to 
learn* lest we should be shamed by contrast." 

2. By supplying good books and magazines 85 

3. General encouragement, by being interested in his inter- 
ests, by praising good work, etc 65 

4. Through directed observation 40 

5. By providing materials for educational play, such as kin- 
dergarten supplies, mechanical toys, wireless appara- 
tus, etc 21 

6. "Have not stimulated, 5 * no reason given 20 

7. By encouraging emulation 14 

8. By endeavoring to form high ideals 10 

9. Have avoided stimulation, afraid of overstrain .... 9 

10. Direct instruction by some member of the household in 
reading, writing, etc 8 

11. Through rewards for school accomplishment .... 7 

12. Encourage work along line of special ability .... 6 

13. By example of parent 4 

14. By awakening curiosity 4 

C. Other matters you have deemed important: 

1. Truthfulness and honesty 66 

2. Companionship between parent and child 43 

3. Religious training 41 

4. Obedience 35 

5. Formation of good social and moral habits 35 

6. Careful attention to health 31 

7. Good reading matter 21 

(Note also that 85 parents have mentioned reading under 
Part B.) 

8. Spirit of cooperativeness, fair play, etc 19 

9. Self-reliance 15 

10. Contact with aesthetic stimuli 13 

11. Instruction in sex matters 13 

12. Cooperation between home and school 11 

13. Logical motivation of conduct 11 

14. Correct habits of speech 10 

15. Careful choice of associates and playmates 10 

16. Outdoor play 9 



EDUCATIONAL HISTORY 285 

TABLE 106 Concluded 

17. Travel 7 

18. To develop spirit of trustworthiness and honor ... 7 

19. Punctuality 6 

20. Habits of thoroughness 6 

21. Concentration 6 

22. Fair-mindedness. Freedom from bias or dogma ... 5 

23. Wide general information 5 

24. Courage 5 

25. Avoidance of fatigue 5 

26. Educational movies, lectures, etc 4 

27. Practical training in household affairs 4 

28. Cheerfulness 3 

29. Develop individuality 3 

30. To see that first impressions are correct 3 

31. Encourage self-expression 2 

32. To keep child modest and free from conceit .... 2 

33. Encourage original research 2 

34. Pre-natal influence 1 

35. To teach dangers of alcohol and tobacco 1 

SUMMARY AND CONCLUSIONS 

1. The average gifted child enters school (above the 
kindergarten) at Q 1 /^ years. Three-fifths have previously at- 
tended a kindergarten, the average length of such attend- 
ance being a little more than half a school year. Low first 
grade is skipped by 21 per cent of the children, and the en- 
tire first grade by 10 per cent. 

2. According to the usual standards, about 85 per cent of 
the gifted children are accelerated and not one is retarded. 
The average progress quotient is 114, which means that the 
average gifted child is accelerated 14 per cent of his age. 
The average P.Q. is somewhat higher than this for the 
younger children and somewhat lower for the older. Ap- 
proximately 85 per cent have skipped one or more half 
grades, as compared with 4 per cent who have repeated. 
The mean net gain from skipping is one full grade. Accord- 
ing to the testimony of the teachers the average gifted child 
merits additional promotion to the extent of 1.3 half grades, 
or about two-thirds of a grade. In all, 82 per cent are said to 
merit some additional promotion. 

3. The discrepancy between mean age and mean mental 
age of gifted children in the first grade is 2.8 years, and by 
grade 5 this is increased to nearly 5 years. 



286 TRAITS OF GIFTED CHILDREN 

4. Teachers' ratings of school work show that the gifted 
children, as a rule, are doing work of superior quality in the 
grade where they are located. The superiority is greatest in 
"thought" subjects and is near zero in such subjects as pen- 
manship, sewing, manual training, games and sports. 

5. Two and a half times as many gifted as control chil- 
dren are rated as "very even" in mental ability, but twice as 
many of the gifted group are rated as "very uneven." The 
girls are rated as uneven somewhat less often than the boys. 

6. Our gifted children have on the average attended two 
different schools by the age of 8 years, and three by the age 
of 11. 

7. The mean estimated number of days of absence from 
school during a year is 12. 

8. Only 1 per cent of the gifted are reported by parents as 
having a positive dislike for school. With 4 per cent more 
the liking is "slight." It is "very strong" with 54 per cent of 
boys and 70 per cent of girls. 

9. According to the school reports, less than half as many 
gifted .as control children display an undesirable attitude 
toward school. 

10. Nearly half of the gifted children learned to read 
before starting to school. At least 20 per cent (and probably 
considerably more) learned to read before the age of 5 years, 
at least 6 per cent before 4, and at least 1.6 per cent before 3. 
Most of these learned to read with little or no formal in- 
struction. 

11. Roughly, 70 per cent of the parents say they have al- 
lowed the child to go his own pace in school, 20 per cent 
have encouraged rapid progress, and 10 per cent have held 
the child back. 

12. The average gifted child does about two hours of 
home work per week on school lessons. Somewhat more 
than half the group have private lessons in such special 
subjects as music, drawing, painting, dancing, language, 
etc. Those who take such lessons devote an average of six 
and a half hours a week to them, including practice. 

13. Parents reported indications of superior ability in 
arithmetic in the case of almost half of the group. The pro- 
portion was about a third for music, and somewhat less for 
dramatics and drawing or painting. The parents of 8 per 



EDUCATIONAL HISTORY 287 

cent report that they had never observed in their children 
any indications of superior general intelligence. For the 
others, the average age at which intellectual superiority was 
first noted was about three and a half years (3.7 for boys 
and 3.25 for girls) . Musical ability first appeared at an aver- 
age age of 5 and the other special abilities at an average age 
of 6. 

14. The indications of superior intelligence most often 
noted were quick understanding, insatiable curiosity, ex- 
tensive information, retentive memory, early speech, un- 
usual vocabulary, etc. 

15. Very few of the parents have carried out any system- 
atic scheme of child training, but a majority have encour- 
aged the child by answering his questions and taking an 
interest in the things which concerned him. 

16. Although the home environment of the gifted chil- 
dren has been, on the whole, above the average, nothing has 
been found to warrant the belief that the superior intellec- 
tual attainments of our gifted group are in any considerable 
degree the product of artificial stimulation or forced cul- 
ture. 



CHAPTER XI 

TESTS OF SCHOOL ACCOMPLISHMENT AND OF 
GENERAL INFORMATION 1 

School progress, estimates of the quality of school work, 
and data on educational history have been treated in the 
preceding chapter. The present chapter gives the results of 
the more exact measurement of educational accomplish- 
ment secured by the use of the Stanford Achievement Tests 
and of a test of general information in (1) elementary sci- 
ence, hygiene, and geography; (2) language and literature; 
(3) history and civics; and (4) the arts. 

STANFORD ACHIEVEMENT TESTS 

Immediately after the search for subjects was completed 
(May and June, 1922) the children of the main experimen- 
tal group were brought together in groups of 10 to 40 and 
given the Stanford Achievement Tests. The number tested 
at each age was as follows, reckoning age as of date the 
tests were given : 





Boys 


Girls 


Total 


6 and 7 


27 


27 


54 


8 


39 


33 


72 


9 


47 


45 


92 


10 


70 


45 


115 


11 


68 


52 


120 


12 


40 


40 


80 


13 and 14 


16 


16 


32 



Total 307 258 565 

Instead of the two information tests of the Stanford 
Achievement battery, a more extended information test of 
the same general type was given. The other achievement 
tests were those of the Stanford Achievement battery un- 
modified. The derivation of these tests, and the procedure in 
giving them, are described in the Manual of Directions. 3 

^Written with the assistance of A. S. Rauhenheimer and Florence L. 
Goodencmgh. 

'World Book Company, 1923. 

289 



290 TRAITS OF GIFTED CHILDREN 

The occasion for undertaking the preparation of Stan- 
ford Achievement Tests was the need for a more reliable 
measure of the educational abilities of the gifted children 
than could be secured by any of the tests already available. 
It seemed a matter of considerable importance to determine 
with a small probable error the educational accomplish- 
ments not only of the group as a whole, but of each individ- 
ual child. Before intelligent provision can be made for the 
education of gifted children it is necessary to know what 
their accomplishments are under the present system: 
whether they are achieving to the limit of their abilities; 
whether they are being unduly retarded; whether their 
accomplishment is better in some subjects than in others; 
whether they show an excessive tendency to unevenness, etc. 
That the Stanford Achievement Tests are capable of answer- 
ing these questions with a rather high degree of accuracy, is 
evident from the reliability coefficients of the separate tests 
composing the battery. These have been computed for the 
separate age groups of 7 to 15 years. The average age re- 
liabilities are as follows : 

Paragraph Meaning .91 

Sentence Meaning .89 

Word Meaning .95 

Computation .87 

Arithmetical Reasoning .88 

Language Usage .80 

Spelling .91 

Educational Age .98 

The average P.E. of an educational age based upon the 
Stanford battery (1.77 months) is so low that the score 
earned by an individual child may be accepted with consid- 
erable confidence. 

The total working time for the separate tests is as fol- 
lows : 



Grades 2 and 3 
Word Meaning 5 min. 
Sentence Meaning 5 " 
Paragraph Meaning 15 " 
Computation 10 " 
Reasoning 10 " 


Grades 4 
10m 
10 
20 
20 
20 


to 8 

in. 


Language Usage 


(Not given in these grades) 


8 




Spelling 


18 " 


(approx.) 


15 


(approx.) 


Total 


63 " 


(approx.) 


103 " (approx.) 



TESTS OF SCHOOL ACCOMPLISHMENT 291 

The time allotments are liberal enough to make each test 
almost entirely a test of "power" rather than of speed. The 
tests measure the child's actual mastery of the various 
school subjects, not his mental agility or his previous ac- 
quaintance with pencil-and-paper tests. 

The control group in this case consisted of approximately 
1,800 unselected children who were tested in the derivation 
of the norms published in the 1923 edition of the Manual of 
Directions. These were enrolled in the public schools of San 
Jose and several other representative small cities of Califor- 
nia. The norms yielded by this group differed little from the 
revised norms of 1924 based upon the analysis of 10,000 
tests from representative cities of the United States. 

Mean scores and mean "subject quotients" have been 
computed for each age and sex group (Tables 107, 108, and 
109). The subject quotients are in each case the quotient of 
the child's age score divided by his chronological age. The 
age norms are those given in the 1923 Manual of Directions. 

In Figure lOa are shown the subject quotient profiles of 
the gifted boys and girls who took the achievement and in- 
formation tests. These are based on the mean quotients for 
the ages combined. For purposes of comparison the mean 
IQ's of the same children who took the achievement and in- 
formation tests are also shown in Figure lOa. 

Figure lOa and Tables 107, 108, and 109 show a re- 
markable superiority of the gifted group over the unse- 
lected school children in educational achievement. The 
mean subject quotients for the different ages fall chiefly 
between 130 and 150. Table 109 gives the distributions and 
mean subject quotients for the ages combined, together 
with the distribution and mean of the corrected IQ's of the 
same children who took the achievement tests. The means 
are as follows, in order of magnitude: 

Boys Girls 

Mean IQ 151.6 151.6 

" Lang. Q. 146.2 148.3 

" Read. Q. 145.3 144.7 

" Arith. Q. 138.5 135.7 

" Spelling Q. 140.2 137.7 

Roughly, the subject quotients are about four-fifths as 
superior as the IQ's. They would doubtless be higher but 
for the fact that promotions earned have been denied. That 



292 



TRAITS OF GIFTED CHILDREN 



O 

< 



S 
S 



CO 



CO 

o 



e 



sl 



i s 



s 



3* 

i? 



O 



O 



, COCO -*rH rH O *,rH CO O 



3 S 



-* o oo j 



* 10 oa co 10 co 

cp* oo * co CO 

g 8 JO S g r, 



O CO-* OS C9 ^*^ *? ' 



C rH O 



OO CO CO 



05 rH 10 t- <NIO CO 10 



CO XO CO 10 O 10 CO 



*# 00 10 



CO CO OS CO ^ 00 

oo TH "* SS W 53 
rH Jt> rH 00 rH O ' 



o O 



C9rH * CO <^<rH CO C^ ^ ^ 



O 10 O rH CO 



S g 3 ft g ft 



rH <O t- OO O 



a s 



* 



B 



oo wj 1 ^ coco e ^'* > ; corn 



GO CO IO t- CO OOC9 CO O CO 



10 co 

s 



IO CO O9 CO O9 CO IO 
4 J> O Jt> O CA rH 
rH 00 rH OS rH OS rH 



CM CO CO J> 



rH CCS CO Jt 

S ^ gs ** 



OJ 00 CO 



CO CO 00 



oo e> co 



OS OS 00 04 



; ft a ft 

OQ OQ 



.ft s ft a ft 

QQ 05 DQ 



a ft 
s 

a 

03 

53 



I 

o 

I 



I 



a 



.13 

o,a 



i 
" 






TESTS OF SCHOOL ACCOMPLISHMENT 



293 



the subject-matter should have been mastered to a point 
more than 40 per cent above chronological age, while the 
children have been held on the average to a grade location 
only 14 per cent above the norm for their chronological 
ages, is a noteworthy achievement. 

FIGURE lOa 
SUBJECT QUOTIENT PROFILE OF GIFTED CHILDREN 

(Based on mean quotients for the ages combined) 



160 




There are only small sex differences in the subject-mat- 
ter achievement of these gifted children, although the boys 
of 9 years and above are somewhat superior to the girls in 
arithmetic, while the girls of 10 and above are slightly su- 
perior to boys in language usage. In both sexes the mean 
subject quotients all increase for a year or two after age 7 
and show a marked decline at 12 and 13. The relatively low 
quotients at 6 and 7 may be due to the fact that the children 
of this age have attended school for only a brief period* The 
decline at 12 and 13 may be due to two causes: (1) the fact 
that the brightest children of these ages were missed be- 
cause of promotion to high school; and (2) the inadequacy 
of the (projected) age norms above 14 years. 



294 TRAITS OF GIFTED CHILDREN 

In general, the subject quotients of a given child tended 
to fun fairly even. Occasionally, however, specialization of 
considerable magnitude was found, and in a good many 
cases the degree of specialization appeared large enough to 
be probably significant. The data bearing on this question 
have been analyzed elsewhere by more accurate methods 
than the use of subject quotients made possible. (See Chap- 
ter XII.) 

TABLE 108 
MEANS AND STANDARD DEVIATIONS OF SUBJECT QUOTIENTS BY AGE 



Means and 
S.D's 


Reading 
Total 
Boys Girls 


Arithmetic 
Total 
Boys Girls 


Language 
Usage 
Boys Girls 


Spelling 
Boys Girls 


Age 


6 and 


7,M. 
S.D. 


139 

15.95 


142 
15.25 


130 
14.47 


133 
10.50 


143 
16.55 


156 
24.70 


136 
13.05 


141 
9.35 


Age 


8, 


M 
S.D. 


151 
13.21 


148 
17.65 


138 
11.36 


140 
14.85 


160 
20.44 


163 
18.35 


145 
12.85 


144 
12.25 


Age 


9, 


M 151 
S. D. 12.63 


151 
12.15 


141 
17.36 


137 
12.35 


159 
13.73 


153 
14.45 


147 
18.85 


138 
7.45 


Age 


10, 


M 
S.D. 


148 
11.23 


148 
8.80 


142 
14.52 


138 
8.55 


147 
14.42 


151 
12.55 


137 
12.38 


137 
12.62 


Age 


11, 


M 
S.D. 


145 
8.02 


145 
9.05 


139 
11.43 


137 
10.10 


141 
11.17 


147 
9.15 


139 
13.97 


140 
12.91 


Age 


12, 


M 
S.D. 


138 
6.69 


137 
6.05 


136 
8.06 


131 
10.40 


133 
9.38 


135 
8.30 


139 
8.6 


137 
11.37 


Age 


13 and 


14,M. 
S.D. 


131 
4.17 


131 
4.95 


131 
6.75 


124 
10.85 


128 
6.75 


132 
6.20 


134 
9.95 


132 
14.45 



It must not be supposed, however, that all of our gifted 
children are satisfactory pupils. The large majority of them 
are, according to the statements of their teachers, but there 
are occasional exceptions. A few are rated low in quality of 
school work because of flagrant neglect of their daily tasks. 
There is reason to believe that others suffer in their class 
marks because of traits of personality which irritate the 
teacher or lead to an unjust appraisal of subject matter 
accomplishment. Gifted children, like others, differ greatly 
in temperament, ambition, personal attractiveness, and 
ability to display their knowledge to advantage. Some lack 
interest because the work is too easy for them. 



TESTS OF SCHOOL ACCOMPLISHMENT 



295 



TH CO Tf O 



8 5 



TH O M lO I> O5 O I> ^ (35 TH 00 TH l> O 00 CO 



CO 00 00 00 O5 IQ Cft 00 t>- CO C^l CO ^ 



Q a- 



ift O lO O 
Oi Gft 00 00 



co o 



CO TH O ' 
CM LO TH 



00 1C 

TH 

1C* 00 

CM 00 TH 
TH 

l> OS 

O> O <^ 

CM -^ TH 

TH 

CM 00 

CO 00 id 

TH 

00 CM 

00 CO* I> 
~1 ^ TH 

TH 

00 
00^.. 

CM 00 TH 

TH 

CMO 

TH 00 00 

00 00 TH 
TH 

00 TH 

XO ^ CM 

CM ^ TH 

TH 

O 00 
O iTi CM 

CO ^ TH 



296 TRAITS OF GIFTED CHILDREN 

GENERAL INFORMATION TESTS 

The general information test 1 used is in part identical 
with the information tests of the Stanford Achievement 
battery, but is more reliable and covers a wider range. Two 
comparable forms were developed, each containing 335 
items, distributed as follows: 

Geography, hygiene, and elementary science 110 items 

Language and literature 90 " 

History and civics 90 " 

The arts 45 

Total 335 

The 770 items included in the two forms were selected 
from approximately 2,500 which were prepared. The 2,500 
items were based upon an analysis of (1) textbooks used in 
grades 3 to 12; (2) courses of study; (3) published studies 
on curriculum content in elementary and high schools; and 
(4) other information tests. Considerable non-scholastic 
material was also included. The intention was to secure a 
reliable measure of general information which would be 
applicable to children of grades 3 to 12. First, the number of 
items was reduced to 1,647 by the combined rating of five 
judges. The 1,647 retained were then broken up into three 
comparable forms, printed, and given to children in grades 
3 to 9 in San Jose, Redwood City, and San Mateo, California. 
Each item of the 1,647 was validated by computing the per 
cent of successful responses in each grade. Individual 
curves were plotted and all items not showing a significant 
and consistent increase in the per cent of correct responses 
from grade to grade were eliminated. These curves were 
based upon fifty pupils of approximately average ability se- 
lected as follows from the entire number tested: (1) Each 
teacher ranked all the pupils of her class in order of general 
scholastic ability; the median pupil, plus the eight pupils 
just above the median and the seven just below were taken 
as representing that grade in that city. (2) Three such 
groups, one from each of the cities in which the preliminary 
tests were given, were combined to represent the grade in 

x Valuahle assistance was given by Dr. G. M. Ruch in the collection of 
material for this test. 



TESTS OF SCHOOL ACCOMPLISHMENT 297 

question. Three cities were used in order to guard against 
the influence of local experience and local instruction. Using 
the per cent data as a basis for the evaluation of the 1,647 
items, two final forms were made up, each having 335 items. 
Each item is a statement containing three response words 
or phrases, the correct one to be designated by underlining. 
In assembling the two forms, care was taken to make them 
analogous in content and equally difficult at all points. The 
individual items were arranged in order of difficulty. Illus- 
trations of items retained in Form B are as follows : 

1. Geography, hygiene and elementary science. 

2. The earth is shaped most like a BASEBALL FOOTBALL PEAR. 
22. The anvil is used by BLACKSMITHS CARPENTERS PAINTERS. 

41. The house-fly spreads BUBONIC PLAGUE TYPHOID YELLOW-FEVER. 

60. The ligaments are attached to the BONES INTESTINES STOMACH. 
80. Cumulus refers to CLOUDS ELECTRICITY EROSION. 

100. The cube of 2 is 4 6 8. 

110. Water enters the roots of plants by CAPILLARY OSMOSIS SOLUTION. 

2. Language and literature. 

2. The shepherd boy who became king was DAVID SAUL SOLOMON. 

22. An example of a noun is BIRD FOR SEE. 

44. Barbara Frietchie sympathized with the ENGLISH SOUTH UNION. 

61. "Styx" was the name of a GIANT GOD RIVER. 
82. E. G. means FOR EXAMPLE SEE BELOW THAT is, 

3. History and civics. 

3. The United States has a KING PRESIDENT EMPEROR. 

23. Roger Williams was COLONIZER JUDGE MERCHANT. 

42. The power of declaring war is vested in CONGRESS PRESIDENT SECRETARY OF WAR. 
60. The ''Invincible Armada" belonged to FRANCE ROME SPAIN. 

82. The Southern States seceded in 1850 1861 1865. 

90. The law of gravitation was first stated by COPERNICUS GALILEO NEWTON. 

4. The arts. 

4. Crayons are used in MODELING DRAWING MUSIC. 
8. A duet is sung by TWO FOUR six. 

22. R. S. V. P. means COLLECT ON DELIVERY INFORMAL REPLY EXPECTED. 
39. Handel is known as a MUSICAL COMPOSER ORGANIST VIOLINIST. 
44. Rodin is famous as an ARCHITECT A PAINTER A SCULPTOR. 

The items in each part of the test cover an extremely 
wide range. For example, in Form B, Test 1 calls for in- 
formation concerning soap, ivory, planes, the burro, tad- 
poles, peat, cloths, the largest state in the Union, soda, 3.1416, 
dynamos, the tides, plant "breathing," insulating materials, 
pollination, etc.; Test 2, concerning Black Beauty, Cinder- 
ella's coach, Hiawatha, Huckleberry Finn's chum, adjec- 
tives, the plural of was, Sir Launfal, gnomes, Vulcan, the 
author of The Raven, the prefix inter, The Jungle Book, 
Milton, etc.; Test 3, concerning Columbus, colonial settlers, 



298 TRAITS OF GIFTED CHILDREN 

Mohammed, the trial by jury, the Red Cross, the Pope, the 
allies of Germany, the feudal system, federal authority, the 
Soviet, Horace Mann, Pericles, etc.; Test 4, concerning movie 
stars, musical notation, mixing paints, jazz, Beethoven, so- 
cial form, stucco, sculpture, architecture, operas, etc. 

One form of the test requires approximately an hour for 
administration. It is given as a "power" test, enough time 
being allowed for all of the pupils to finish. Score is the 
number of right responses minus half the number of wrong 
responses. 

Both forms of the test were given to 463 unselected pupils 
(216 boys and 247 girls) fairly evenly distributed in grades 
3 to 9 in the cities of San Jose, Sunnyvale, and Mountain 
View (California) for purposes of standardization. These 
pupils served as the control group in the evaluation of the 
scores earned by the gifted children. Half of the control 
group took Form A first; the other half, Form B first. 

Reliabilities were computed by age and grade by corre- 
lating the scores of Form A against those of Form B. These 
are given in Table 110. 

TABLE 110 

RELIABILITY COEFFICIENTS OF THE INFORMATION TEST 

r 
Grades Form A vs. Form B S.D. of Form A S.D. of Form B 

3 .60 14.3 13.2 

4 .79 26.3 26.8 

5 .84 34.0 29.6 

6 .87 41.8 40.6 

7 .95 47.7 49.1 

8 .90 40.0 41.1 

9 .88 43.0 39.9 

Ages 

8 .85 17.6 17.3 

9 .89 30.3 30.5 

10 .94 35.6 37.5 

11 .89 54.4 53.1 

12 .96 69.9 61.2 

13 .91 68.5 66.9 

14 .94 65.8 62.4 

15 .95 72.1 72.7 

The average reliability for a single age group is .917. The 
reliability of two forms, computed by Brown's formula, is 
.96. Only one form was given to the gifted group. 



TESTS OF SCHOOL ACCOMPLISHMENT 299 

Age norms were derived for the total score and for each 
of the four tests by smoothing the age means. In each case 
this gave approximately a straight line from age 8 to age 15. 
In order to make possible the calculation of information 
quotients for the brighter children, the line was extended 
beyond 15. The "age" norms above 15 are thus fictitious 
units, equal in size to the units in the lower range. This is 
not a refined statistical procedure, but it affords a conven- 
ient method of comparing high scores when minute exact- 
ness is not demanded. 

Means and S.D's of the gifted and control group by age 
and sex are shown in Table 111. 

Table 112 gives the distribution of information quotients 
of the gifted group by age and sex separately and for ages 
and sexes combined. 

The superiority of the gifted group in general informa- 
tion is seen to be very great indeed. This holds for both 
sexes, at all ages, and for each of the four types of informa- 
tion tested. In most of the comparisons the gifted excel the 
control by from two to five times the S.D. of the latter; more 
commonly by three or four times. The mean information 
quotient is about as high as the mean intelligence quotient, 
and is somewhat higher for gifted boys than for gifted girls. 
This sex difference is not found with the control. At ages 
8, 9, and 10, not a single child of the control group reached 
the mean score of the gifted of corresponding age. The 
mean of the gifted boys was reached by 3.6 per cent of the 
control boys at age 12, and by 2.2 per cent at age 13. The 
mean of the gifted girls was reached by 2.5 per cent of the 
control girls at age 11 and by 7.7 per cent at age 12. Only 2 
of 291 gifted boys, and 5 of 242 gifted girls, earned an in- 
formation quotient below 120. Not a single gifted child fell 
as low as the average, of the control. The difference between 
the two groups is somewhat less in history and civics than in 
the other types of information tested. 

Table 113 shows that the mean quotients of the gifted 
children tend to run highest for language and literature, 
and lowest for history and civics. Only in the arts score do 
girls equal the boys. In science and history the boys excel 
the girls by a considerable margin. 



300 TRAITS OF GIFTED CHILDREN 



o 



II 



o o CM os oq c> r^e^jt^ i>; os oq H; oq CM co t>; cq 
" 



g TH os 06 rH TH CM" eo ed o OCMOO oo 10 JH 

- - 



cooor- *- 



I 

H 13 00 TH 00 _ .. _... 

TH CO 03 t- xo CO O OS t- * O ^ 



OS O US CO OS O OS CO OO Oft CM ^ r-t CM F>; "* O OO 
CO IO CO CD liO ** CD "* lO OS OS O 1 *^ CO CO CD t 1ft 



g o 



Q^^OOSrHiOXO-^HOOxOOqcOCM 1 

O 5n OS OO OS ^ >* W* OS TH !>I -^ ^ ^H OS OS OS CM 1-1 CO D-* CO OO 
^^ r*3 | T4 T-^ f4 CM rH CM CM CM CM CM CM CO CO CO CO CO CQ 

O 



( w -^ r^t^OSrHOOSOOvOC^JO^OOSOSOOSCMOS 

J IS co^oi r^'oSuS rncoos- oogS S2S SS 

3 

w a 

O ^Jaj T(COO OOCMOS CO T < lO COxOOft C^C**CO r-C T-I IOO<O 



--<U T-JCOO OOCMOS 

SH ^^loO rHlOC CMC^-r- !>-THiO THiOio OSCMt->. lOCOlO 
g rHrH CMCMrH COCOCM ^\O-* CO^ " " 



_ . .... CMCMTH COCOCM ^\0-* CO CO kQ O t> CO r- 

5| _ 

15 P ^ooTHoqos^^coTH^ioocN|r^w3 < *;oco'* 

OOi-Ht^ COCM^* iO*OA 5253^ CMC51O f-iOOO 



a 



OS CO CD t 1 * CM OS O CO O "^ ^* GO t* rH CO CM ^f OO CO CM CO 

tiT Sc rHCMrH ^I>^TH lOc^cM CO-^TH lor^co CM'^-HCM* c^cooo 
M HH--5 CMCMCM cococo ^rH'rt* -^lOkjO cDcoco r>-r r tr>-i> 



CO ^ C^ * ^ ^ ^> ^ **? 9 9 TH O kO Tj; CO O CM 

fa e$ G CO OS kO lO kO XCD rH CM OS OQ G> f* OJI CO O t^ C^ CO 



"S * OO CO rH O tq CO O O OO CO Oft CO ^ CO O lA f; OS iO OO -^ 
M CM ^H rH CS CO CO Tf CM -^ CM * OS CO CD OS O ^ CO D^ O ^ 

^t>-5O OOOOC^ OOOiOO 



CQO CQO MCJ CQO PQ O PQ O 



TESTS OF SCHOOL ACCOMPLISHMENT 301 

2 OOTHC 

*- r- CO T~ 



i to oq CM ^ CM oq r- oq ri< -^ co CM 

S i> CO OS Cft-rHt^ CO OO* CM* vOOO l3 TH o OS TH xft 

O ;j TH CM CMCOCMCOCOCOtfSCOtOCOCOr-COt'-CO 

S/2 

1 

E" 1 F- CM \0 S CO S C^ 00 ? 



o rH Cft COr-^CS TH OO* lS CO rH -r * CM CO 

CO CO d COCOC? T* CO CO CO 00 CO CO CO CO C3 <M GNI r-* t-< r-1 



I 

O 2 THiiaor^oo^THOOcoTHcoxrtooocococor^ 



O .- CM > ^CMCOCOCO'!l l| ^t<*COcOCOOOOOOOC*C 1 SOO 

g5 u 

S e 

^ IH o oa c^ * co c^ CM os TH co co oo oo cococo TH cs CM o t- oo 

Q ^- COCO"* COCMCM k5I>oq COC^T^ COC^r^- OOCaoq OOCOTH 

Q 
S5 

<J 

W tj va 40 TH co oo o co co CM co r- -^ f- co <z> r- t^ ^ 

c* M "S ^^^^^^^ o * I> ; cs ? c ^| c '? e ' r 5 t<< ! C3 * cs 2^'^*I 



arh^CJ THTH THTHTHCMCMTHCMCMTH 

CJ O 

G co" TJ 

cj w a 

o (fi ctf 

- w Q . T3 COCOCO f^-CMCO CTH-^ COrH<o CftCftOO 

^ S3 5.H JS^TH S^S SSS THTHTH 

3 2 

CO !^ "a OO^OOOTH^t'-THCOOCOOOOascMCMOO 

-5 P5 S co oq CJM oo TJJ os oq oq iq as TH co co \a \q TH co TH 

f"H P*Md kOCO^OSOr^CSOOOiOCOiAodlOOOOaO 



TH !>. CM D tt CO r- 00 CM CO OS OS OO OO l> "* CO TH 00 

>; cooqc*; coioos cooc^ CO^CM cococo 0005^ 
ooos ooos ,-^THTH OCSTH r^c^t>" - - ' 



Q 

Jz; 

" TH CM co i>-osi>- CM o a* CM t> 

"* 



d COOO^* COCOCO -^lCM CSTHt THCOOS THCJJO 



.% ^ cocoas co r- ^? ^rr'Os TH TH oo ipcooo op co oo o o co 



oq oq -^* CMCOCO oq rjj i>- \o o oq 

TH TH CM TH ci OS ^ CM CM TH O CM CM TH TH OS 00 OS OO r- OO 



tv- 

S&s &s c?s &s &s c?s &S 

P5O MO PQO PQO PQO PQ O WO 
rt o T-. CM co 

COOOOSTHTHTHTH 



302 TRAITS OF GIFTED CHILDREN 



a 

3 






5 

s g 

H 



r^ 2 



CO ^ > 



- (MOOOC* ;g 



2 rH O> OO <M CO CO 

8 

g ^ CO O rH O CO T* 

O 
O ^^>01>100^< 



03 THOQiOiOOOl>OOOiC&lOtH 



es t7 co 

fe O Tf< -TH 

*-* o ^ 

t^*r OO 

P2 lO rH 

Jgr rH 

2 o 

P Cs 'rf -rH 

2 00 ** 

I rH ^ ^ 

Q rH ^ 

CO 

^ rH C^|COTHrHd<MdCOTH OCO 

t* Tf* rH 

-O I- 4 

fl ^ 

<e M rHCVlrHrHCOGMCSICOrHrH iO-^ 



TESTS OF SCHOOL ACCOMPLISHMENT 



303 



O CO 



CM CM 

10 



CO CO ^i* 





I 



(g CO 






H 



pS 
B 



I S 
S S 



o ^ 



CM 

10 



CM 

iO 



g TH OT oo c^ oo 

^ TH TH TH TH TH 



2 w 

^ r/5 w 



CO 

TH ^ 

H W 

S w 

S 

5z; 

I 



73 CM 

3 S 



o ^ 

tfl TH* 



"C 
3 S 



00 

O 



10 10 



O IO lO ^ iO 

CQ TH TH TH TH 







^ 3 2 



-f CM 






C/5 

? - 



4) ^ 

o *y 

O ti 

o e ^~* 

"3 | ^2 

CO ^ W 



CO 

"3 



a; 



<D 



Q 

00 



> 

" oo 



.2 
"5 



TH ^ CM ... 

S 3 S ^ 

J3 

<O TH TH fl 



A IO 



?? 
< ^ 



CO 
10 



iO 

CO 



CM CO 



4H 05 



a 
8 



S CM 



* | CO 

& 



t; as 

S CM 



CO 



10 

CO 



o ro 

^ II 

50 3 

I % 

o o 

? ^ 



a; y 

_ 






0) 

1 



2 

s 

a 



304 TRAITS OF GIFTED CHILDREN 

That the gifted make a showing in general information 
even better than they make in such subjects as reading, 
arithmetic, and spelling, is probably due to the fact that a 
child's stock of information is more dependent upon intel- 
lectual initiative and less upon formal school instruction. It 
would seem, therefore, that general information tests might 
be more valid than achievement tests for use in selecting 
pupils for special instruction in gifted classes. They prob- 
ably compare well with the best group intelligence tests for 
this purpose, and they have the advantage of being easier to 
administer. In giving an information test of the kind we 
have used, it is only necessary to place a blank before the 
child and let him work through it. No supervision is neces- 
sary except to see that aid is not secured from books or 
from other pupils. If the child cannot be brought for a test 
to an examiner who is familiar with test procedure, a blank 
may be sent for use by any teacher or parent, with no risk 
that the score will be invalidated by incorrect procedure. 

THE INFLUENCE OF ATTENDANCE UPON EDUCATIONAL 
ACCOMPLISHMENT 

The outstanding fact is that these children have accom- 
plished so much in the various school subjects with such a 
brief period of instruction. For example, a gifted child of 8 
who has attended school only two years has usually mas- 
tered the work of about five grades as well as it has been 
mastered by the average unselected cfyild at the time he is 
considered ready for promotion to the sixth grade. A gifted 
child of 7 who has attended school one year has usually 
mastered the work of three or four grades. Actual length of 
school attendance seems to play little part in determining 
accomplishment in comparison with the part played by na- 
tive endowment. 

It is possible to secure a measure of the influence of 
length of school attendance upon educational accomplish- 
ment. Possible determinants of accomplishment are age, 
intelligence, and length of school attendance. The influence 
of any two of these may be eliminated by means of the 
partial correlation method or by treating the data of a 
group of subjects who have the same (or practically the 
same) chronological age. We have taken for treatment the 



TESTS OF SCHOOL ACCOMPLISHMENT 305 

109 cases having a chronological age of 10 to 11 years at the 
time the achievement test was given. The IQ's of these chil- 
dren ranged from 139 to 190, with S.D. of 10.44. The length 
of school attendance ranged from 2 years to 6% years, with 
S.D. of .858 of a year. 

The following correlations were found: 

School attendance vs. Spelling Q, r =.041 .066 
Information Q, r =.003 .067 
Reading Q, r = -.099 .066 
Arithmetic Q, r =.132 .065 
IQ, r -.013 .067 

IQ vs. Spelling Q, r -.328 .060 
Information Q, r =.457 .052 
Reading Q, r =.342 .059 
Arithmetic Q, r -.261 .062 

It is evident that school attendance, although it has 
varied greatly, has had no appreciable effect on subject 
matter accomplishment On the other hand, the IQ, al- 
though the range is relatively very small, has been an im- 
portant factor in determining accomplishment. 

SUMMARY 

The educational accomplishment of 543 children of the 
main experimental group was measured by the Stanford 
Achievement Tests in reading, arithmetic, language usage, 
and spelling, and by a general information test of 335 items. 
The results on all of these tests have been compared with the 
test scores of unselected school children of corresponding 
age. The most important findings are as follows : 

1. The superiority of the gifted children of a given age 
over unselected children of corresponding age is very great, 
amounting in most cases to from three to four times the 
S.D. of the unselected age group. This superiority holds for 
all the fields of accomplishment tested, at all ages, and with 
both sexes. 

2. The accomplishment quotient of the gifted in the vari- 
ous school subjects tends to run from three-fourths to four- 
fifths as far above the average as do the intelligence quo- 
tients. The information quotients, however, run about as 
high as the intelligence quotients. No child in the gifted 
group earned an information quotient as low as 110. 



306 TRAITS OF GIFTED CHILDREN 

3. In general, the average gifted child has mastered the 
subject matter of instruction to a point 40 per cent above 
his chronological age, although he has been held back to a 
grade location only 14 per cent beyond the norm for his 
chronological age. 

4. The superiority of the gifted is greatest in general in- 
formation, language usage, and reading, and least in history 
and civics. The quotients of the gifted are higher in the age 
range 8 to 12 than for younger or older children. 

5. Gifted boys excel gifted girls in general information, 
arithmetic, and spelling. The girls of 10 and above are slightly 
superior to boys in language usage. When the scores of the 
sexes on the separate parts of the information test are com- 
pared, the boys are found markedly superior to the girls in 
science and history and somewhat superior in language and 
literature. On information relating to the arts the girls do 
as well as the boys. 

6. There were few cases of extreme unevenness among 
the various quotients of a given child, but degrees of special- 
ization large enough to be of possible significance were 
fairly numerous. 

7. The accomplishment quotients of a considerable 
number of gifted children are higher than the teachers' 
marks given on the basis of daily performance in the class- 
room would lead one to expect. Presumably, in such cases 
the teacher has either underestimated the child's accom- 
plishment or has given low marks as a penalty for lack of 
application to the set tasks of the school. 

8. At a given age there is practically no correlation be- 
tween educational accomplishment and the number of 
terms the gifted child has attended school. The causes of 
school success and of school failure lie deeper. 

9. The general information test described in this chapter 
is an excellent test for use in the identification of gifted 
children. 



CHAPTER XII 

SPECIALIZATION OF THE ABILITIES OF GIFTED 

CHILDREN 

James C. DeVoss 

THE PROBLEM 

Both vocational and educational guidance must rest 
upon the counsellor's ability to estimate the equipment of 
the one being guided. This estimate must not be limited to 
the general bird's-eye view of the abilities of the child, but 
must include such precise statements as, "This boy is better 
in arithmetic to such and such a degree than he is in compo- 
sition," or "better in history to such and such a degree than 
he is in literature." Since the gifted children represent high 
potentiality, our ability to make precise statements concern- 
ing their equipment is a matter of great consequence. It is 
to supply a basis for deriving such statements that this 
investigation was undertaken. 

In regard to the specialization of abilities several im- 
portant queries arise. Are gifted children more specialized 
than other children in their development? Are they one- 
sided in their development, or do their abilities lie on a 
uniformly high level? When one ability is highly developed, 
is this development at the expense of other abilities? The 
answers to these questions can be best read from precise de- 
scriptions of the surfaces of the abilities of the gifted chil- 
dren. These descriptions necessitate the accurate measure- 
ment of the abilities to be described. 

The abilities under consideration are those which were 
measured by the Stanford Achievement Test, a group of 
four information tests and the Stanford-Binet Scale. This 
measurement is but a small sampling of the abilities of a 
child, but it includes those of considerable importance in 
school work. In the measurement of reading there are tests 
of word meaning, sentence meaning, and paragraph mean- 
ing. For arithmetic there are measures of computation and 

307 



308 TRAITS OF GIFTED CHILDREN 

of reasoning or problem solving. Of the other school sub- 
jects there are measures of Language Usage, of Spelling 
from dictation, and of information in science, in literature, 
in history and civics, and in music and art. 

The use of these tests is best defended by the statistical 
treatment to which they have been subjected as described 
elsewhere 1 and here. Certainly the authors are justified in 
the statement that "these tests have . . . reliability coeffi- 
cients which are very much higher than we are accustomed 
to find in the experimental literature on educational tests." 

METHOD OF TBEATMENT OF DATA 

The data treated in the following pages were secured as 
described on page 310. The arrangement of the data for 
the study of the unevenness of the abilities measured, de- 
manded that the scores be expressed in units common to all 
of the tests. Several such units are available. Quotients, T 
scores, grade percentile scores, and z scores (or standard 
scores) were considered, with the result that the last named 
were adopted. The standard score expresses the distance 
from the mean of an age group in terms of the standard de- 
viation of that group. Thus a standard score of 1.5 in Arith- 
metic Reasoning means that this score is 1.5 standard devia- 
tions above the mean for the age group in which the individ- 
ual making the score was found. 

Use was made of these standard scores in connection 
with four types of study: (1) The application of Kelley's 3 
ratio o&oow This ratio is applied to a gifted and to a control 
group so that the unevenness of the two groups may be com- 
pared. (2) A general survey of the nature of the unevenness 
of the age groups of the gifted children, including some 
comparisons between the various achievement levels and 
the intelligence level as established by the Stanford-Binet 
test. (3) Several studies of types of unevenness of both gifted 
and control groups. (4) Case studies with the assistance of 
profile charts or psychographs. 

Stanford Achievement Tests, Manual of Directions, Kelley, Truman 
L., Ruch, Giles M., and Terman, Lewis M. World Bk. Co., 1923, p. 6. 

3 Kelley, Truman L. "A New Method for Determining the Significance 
of Differences in Intelligence and Achievement Scores." Jour. Educ. Psych. 
Vol. XIV, No. 6, September, 1923, p. 321. 



THE SPECIALIZATION OF ABILITIES 309 

For three of these methods there is no significant histori- 
cal account other than the reference cited. The fourth, rep- 
resenting case studies and profile charts, has a rich historical 
background. The development of the method of case studies 
rests directly on the development of the method of measure- 
ment of human traits. The history of such measurement is 
too long to be recounted here, and it has been often pre- 
sented in textbooks on the subject. The advantages of the 
profile method were first set forth by Rossolimo, 1 who em- 
ployed tests of ten different functions. Each test was scored 
by means of a scale of ten steps, representing just not any 
ability to do the test, and 10 representing perfect perform- 
ance. Claparede 2 has justly criticised Rossolimo's method 
of scoring because of its arbitrary character and has sug- 
gested a profile based upon percentile scores. Bartsch 3 has 
extended the Rossolimo method by the addition of a num- 
ber of tests, but the Rossolimo-Bartsch method still retains 
the weakness of the original method in its arbitrary gradu- 
ation. In America the profile method has been used by 
many investigators. Courtis 4 employed it with his arith- 
metic tests, using scores for points on the profile and both 
grade norms and arbitrary standards for comparison. 
Kelley 5 used percentile charts for eleven dimensions in his 
study of delinquent boys. Several types of profile charts are 
described and illustrated by Hollingworth. 6 These she calls 
psychographs. The Stanford Achievement Test Manual 
gives instructions for constructing a profile based on sub- 
ject ages. These are representative of the profile methods 
that have been used. None of these methods would be as 
suitable for the present purpose as the use of standard 
scores. 

a Rossolimo, G. Die psychologischen Profile. Klinik f. psychische u. 
nervose Krankheiten, VI, 3 u. 4; VII, 1; VIII, 2. 

3 Clapar6de, Edouard. Profiles psychologizes. Archives de Psychologic, 
XVI, 61, pp. 70-81. 

'Bartsch, Karl. Das psychologische Profit. Carl Marfiold Verlagsbuch- 
handlung, Halle an der Saale, 1922. 

4 Courtis, S. A. The Courtis Standard Tests. Detroit, 1914, pp. 45. 

8 Kelley, Truman L. Mental Aspects of Delinquency. University of 
Texas Bulletin, No, 1713; March 1, 1917, pp. 49 and 52. 

'Hollingworth, Leta S. Special Talents and Defects. The Macmillan 
Co., New York, 1923, 



310 TRAITS OF GIFTED CHILDREN 

A COMPARISON OF A GIFTED AND A CONTROL GROUP BY THE USE 
OF KELLEY'S RATIO 

The comparison of gifted with unselected children re- 
quired a selection of two representative groups which should 
have similar opportunities for development of the abilities 
measured. As this study called for two hours of measure- 
ment of each child, and the calculation of 40 or more coeffi- 
cients of correlation for each group, it was fortunate that 
data were already available for a group of 96 unselected 
children (from an eighth grade). 1 It remained to secure 
data from a suitable gifted group. The 643 gifted children 
show a very wide range of grade placement, of chronological 
age, and of mental age. A group of gifted eighth grade chil- 
dren would not be satisfactory because their scores would 
too often be close to or actually at the maximum possible 
with the Stanford Achievement Tests. It therefore seemed 
best to select a group which would represent about the same 
level of achievement as that represented by the eighth grade 
group. After an inspection of groups selected by the use of 
chronological age and grade placement as criteria, these cri- 
teria were abandoned and a group was selected on the basis 
of mental age. The range taken was from M.A. 14-0 to MA. 
15-5. This is roughly eighth grade ability. Inasmuch as the 
mental ability of a child largely determines school success 
and has considerable influence on school progress, this range 
of mental age assures a certain homogeneity. Furthermore 

TABLE 113a 

DATA CONCERNING THE 100 GIFTED CHILDREN SELECTED FOR 

SPECIAL STUDY 

Years in 
Mental Age Chr. Age IQ School Grade School 

Mean 14.7 yrs. 9,86 yrs, 149.4 H. 5th 4.6 

S.D. .44 " .6 " 9.23 1.57 .82 

Range 1.5 " 2.6 " 44.0 5.0 4.5 

Range 14.0-15.4 8.5-11.09 136-180 L.3rd~H.8th 2.5-7.0 

the comparison contemplated in this study depends upon re- 
liability coefficients and intercorrelations and does not rest 
on a comparison of the scores made by individual pupils; 

'For these data the writer is indebted to Dr. Truman L, Kelley. The 
method of treatment is also Kelley's. See "A New Method for Determining 
the Significance of Differences in Intelligence and Achievement Scores." 
Jour. Educ. Psych. Vol. XIV, No, 6, September, 1923, p. 321. 



THE SPECIALIZATION OF ABILITIES 311 

hence the degree of homogeneity of opportunity represented 
by this selection of gifted children is the best obtainable. 
Table 113a gives the important facts concerning this group 
of gifted children. 

In Table 114 are shown the scores made by five of the 
gifted children. This table illustrates the difficulty of judg- 
ing the significance of the differences in abilities when they 
are expressed in terms of scores. The question, is E. A. 
better in Arithmetic Computation or in Arithmetic Reason- 
ing, cannot be answered from this table. However, when 
the scores are given in terms of their distances above or be- 
low their respective age norms, such questions as the one 
asked may be answered. Table 115 gives the means, stand- 
ard deviations, and reliability coefficients of the tests for 
the group of gifted children which was selected for this com- 
parison. 

TABLE 114 
SAMPLE SCORES ON THE TESTS USED 

E. A. A. B. G. A. L. B. E. B. 

Arithmetic Computation 88 100 160 128 96 

Arithmetic Reasoning 48 84 100 108 48 

Word Meaning 55 56 67 77 60 

Sentence Meaning 54 54 78 78 54 

Paragraph Meaning 84 88 94 118 78 

Language Usage 38 47 52 54 49 

Spelling 88 130 174 170 116 

Science Information 46 60 75 81 42 

Lang, and Lit, Information 38 46 44 75 47 

Hist, and Civics Information 19 50 45 64 22 

Music and Art Information 16 25 29 38 21 

Arithmetic Total 136 184 260 236 144 

Reading Total 193 198 239 273 192 

Information Total 119 181 193 258 132 

Using the means and standard deviations of Table 115, 
the raw scores of Table 114 are changed so they state the 
distance which each score is above or below the mean in 
terms of the standard deviation of the distribution of such 
scores for this group. These standard scores for the five 
illustrative cases are given in Table 116. Thus the score of 
88 in Arithmetic Computation for subject E. A. is equivalent 
to the standard score 1.059. (If z is the standard score, 

Score Mean - 88111.7 ^ Acrnv 

z= ~r 3 -TTC r-p Therefore =____= 1 .059) 

Standard Deviation 22.36 



312 



TRAITS OF GIFTED CHILDREN 



TABLE 115 
DATA CONCERNING THE SPECIAL GIFTED GROUP 



Standard 

Mean Deviation 

Arithmetic 

Computation 111.7 22.36 

Arithmetic 

Reasoning 76.71 17.78 

Word Meaning 57.68 9.32 

Sentence Meaning 54.43 11.198 

Paragraph 

Meaning 82.90 10.67 

Language Usage 37.37 8.41 

Spelling 124.80 21.69 

Science 

Information 56.85 13.51 

Language and 

Literature 

Information 47.65 11.99 

History and Civics 

Information 37.40 15.29 

Music and Art 

Information 21.55 6.44 

Arithmetic Total 188,60 35.19 

Reading Total 194.30 28.00 

Information Total 162.4 40.60 



Reliability 
Coefficients 

of the 
Tests for this 

Group 

.898 



.864 
.916 
.836 

.768 
.582 
.952 

.767 

.885 
.863 

.784 
.909 
.924 
.960 



These standard scores for the five subjects are given in 
Table 116. From this table we can answer the question con- 
cerning the interrelationship of the abilities of E. A. He is 
only a very little better in Arithmetic Computation than he 
is in Arithmetic Reasoning; the difference is .551 standard 
scores. Other differences in his abilities are also apparent. 
The difference between his scores in Paragraph Meaning and 
Arithmetic Reasoning is 1.51 standard scores. We shall call 
such differences as these d's. Then d=z^ z 2 , when Zi is the 
standard score in one subject such as Paragraph Meaning 
and z 2 is the standard score in another subject such as 
Arithmetic Reasoning. 



THE SPECIALIZATION OF ABILITIES 313 

TABLE 116 

SCORES OF TABLE 114 EXPRESSED AS STANDARD SCORES 

E. A. A. B. G. A. L. B. E. B. 
Arithmetic 
Computation -1.059 - .52 2.16 .72 - .70 

Arithmetic 

Reasoning -1.61 .41 1.31 1.76 -1.61 

Word 

Meaning - .29 - .18 1.00 2.07 .25 

Sentence 

Meaning - .04 - .03 2.10 2.10 - .04 

Paragraph 

Meaning .1 .48 1.04 3.29 - .46 

Language 

Usage .07 1.14 1.74 1.98 1.37 

Spelling -1.70 - .24 2.26 2.08 - .41 

Science 

Information - .80 .23 1.34 1.79 -1.10 

Language and 

Literature 

Information - .80 - .14 - .30 2.28 - .05 

History and Civics 

Information -1.2 .82 .50 1.74 -1.00 

Music and Art 

Information - .86 .53 1.15 2.55 - .08 

Arithmetic 

Total -1.49 - .13 2.02 1.35 -1.26 

Reading 

Total .04 .16 2.59 2.78 - .08 

Information 

Total -1.07 .46 .75 2.35 - .75 

A striking difference is that for L. B. Her score in Para- 
graph Meaning is 2.57 standard deviations above her score in 
Arithmetic Computation. Is this d a true, valid difference or 
is it a spurious difference occurring because the tests used 
are not perfectly reliable? Certainly if the test in Paragraph 
Meaning and the test in Computation both lack reliability 
the difference of 2.57 standard deviations may be spurious. 
Also the d of .03 between the z score for Word Meaning and 
the z score for Sentence Meaning for subject L. B. would 
seem to indicate that this pupil has practically the same 



314 TRAITS OF GIFTED CHILDREN 

amount of ability in these two fields. If the tests for Sen- 
tence Meaning and Word Meaning were very unreliable a 
real difference might be hidden. 

There is a test of the validity of these differences. Sup- 
pose that we had the true z scores found from the average 
of the z scores secured by taking the test an infinite number 
of times. Call this true z score for Paragraph Meaning Zoo 
and the true z score for Computation z w . 

Then z x z^ the true d between Computation and 
Paragraph Meaning for L. B. What is the probable diver- 
gence of the obtained d's from these true d's? Kelley 1 has 
supplied the standard error for such differences. The for- 
mula is 0^.00 w V2-rii-r sn in which o&oow is the standard de- 
viation of the difference between the true d (z*> -z w ) and the 
successively obtained d's, using different forms of the same 
pair of tests on one individual; r n is the reliability coefficient 
on one test, say, Computation, and r 2U is the reliability co- 
efficient of the other test, such as Paragraph Meaning in the 
case discussed above. 

As Kelley has said, "This formula fills a long felt need, 
since it makes possible the determination of the probable 
errors of our judgments of differences of abilities within the 
individual." 

The P.E. (of individual Zi - z 2 )=.6745 V2 - r iz - r 2ir . 

Table 117 shows the P.E. as calculated by this formula 
for our group of 100 and for all of the possible differences 
in scores. 

Returning to the difference of 2.57 standard deviations 
encountered in the case of L. B. as the difference between 
her z scores in Paragraph Meaning and Computation, we 
can answer the question concerning the probable validity 
of this difference. From Table 117 we find the P.E. of 
this difference is .39. As the difference is 6.59 times as great 
as its P.E. we may conclude that it is a true difference. We 
may thus use the P.E. as a test of the differences encoun- 
tered in the scores of individual children. The method 
opens important possibilities for vocational and educational 
guidance. We have frequently assumed that differences 

iKelley, T. L. A New Method for Determining the Significance of Dif- 
ferences in Intelligence and Achievement Scores. Jr. Ed. Psych. Vol. XIV, 
No. 6, September, 1923, pp. 321-334. 



THE SPECIALIZATION OF ABILITIES 315 

such as these exist, and achievement and educational quo- 
tients have seemed to support the assumption. With this 
formula we may give the assumption statistical precision. 

Again referring to L. B. as an example, we may show 
some of the differences between her scores with probable 
errors, thus: 

Difference, Reading Total - Arithmetic Total = 1.43 P.E.=.28 
Spelling -Arithmetic Total- .73 P.E.= 25 

" Music and Art 

Information - Arithmetic Total = 1.20 P.E.^.37 

" Information 

Total -Arithmetic Total = 1.00 P.E.-.24 

" Music and Art Science 

Information -Information = .76 P.E.=.45 

" Language and History and 

Literature -Civics = .54 P.E.=.34 

Information Information 

" Music and Art History and 

Information -Civics = .81 P.E.=.40 

Information 

Here is the case of a girl 11 years of age in the 7th grade, 
with an M.A. of 15 and an IQ of 142, who is definitely more 
proficient in Reading, in Literature, and in Art than in 
Arithmetic and Science. 

This use of the probable errors may be extended indefi- 
nitely. Inasmuch as there are no probable errors in this table 
larger than .54, and the average of all the probable errors 
shown is .276, and since differences of 1.00 and greater are 
not uncommon in Table 116, there are good prospects for 
the location of many significant differences. 

Before undertaking a detailed study of individuals it will 
be of interest to discover the proportion of cases which pre- 
sent significant differences. The method is also one devised 
by Dr. T. L. Kelley. 1 As pointed out by Kelley, "The fre- 
quency with which differences of various sorts are revealed 
by these fallible tests will not exactly parallel the importance 
of the differences in the natures of the pupils studied, be- 
cause the tests are not equally reliable. Thus if given four 

Kelley, T. L. A New Method for Determining the Significance of Dif- 
ferences in Intelligence and Achievement Scores. Jr. of Ed. Psych., Vol. 
XIV, No. 6, September, 1923, pp. 321-334. 



316 TRAITS OF GIFTED CHILDREN 

TABLE 117 

PROBABLE ERRORS OF THE TRUE DIFFERENCES (P.E. d<oow ) OF 
INDIVIDUAL Z^ Z w2i 



II ll 11 II i\ If ii It !i il I! tl I 





I 



Arith. 
Reas. .329 

Word 

Mean. .291 .316 

Sent. 

Mean. .348 .370 .335 

Par. 

Mean. .390 .409 .379 .425 

Lang. 

Usage .486 .502 .478 .515 .544 

Spell- 

ing .261 .290 .245 .311 .357 .460 

Sei. 

Inf. .390 .409 .380 .425 .460 .544 .357 

Lang. 

& Lit. .314 .338 .300 .356 .397 .492 .272 .398 

Hist. & 

Civics .329 .352 .317 .370 .410 .502 .290 .410 .338 

Music 

& Art .380 .400 .370 .416 .452 .537 .347 .453 .388 .400 

Arith. 

Total .282 .341 .383 .481 .252 ,384 .306 .322 .374 .244 .276 

Read. 

Total .285 .311 .474 .238 .375 .295 .311 .365 .230 

traits, a, 6, c, d, such that children intrinsically vary as much 
in the difference a- & as in the difference c-d, and given 
further measures of a and b which are more reliable than 
those of c and d t then we will be able to discover and deter- 
mine differences a - b more often than differences c-d. We 
must therefore keep in mind that the ease with which differ- 
ences are discovered by the aid of the Stanford Achieve- 
ment Tests depends upon: first, the extent to which the 
individual's abilities differ, and, second, the reliability of 
the tests." 

What proportion of the cases in our special group of 
100 gifted children show differences of the type Zi-z 2 , 
which are so great as to be significant? 



THE SPECIALIZATION OF ABILITIES 317 

As we have seen, the standard deviation for differences 
such as Zi - z 2 for an individual is 0^.00*? =V2 - ru - r 2 n. In the 
case of L.B. we used this (*&<, to test the probable verity of 
the difference between z\ for Paragraph Meaning and z 2 for 
Computation. We found 0d.oo=-58. If the standard devia- 
tion for the 100 similar differences found in our group 
should be the same as this it would be evident that the ob- 
tained differences are not greater than chance would in- 
dicate. 

The usual formula for the standard deviation of a dif- 
ference is at Vovi 2 +tf*2 2 ~ 2r 12 a3ia*2 =V2 - 2r i2 in which Ca 
is the standard deviation for the 100 differences and r 12 is 
the correlation between the distributions of Paragraph 
Meaning and the Computation scores. 

The correlation between the Paragraph Meaning and the 
Computation scores is .25. Then <?<* = V2-2(.25) =1.24. 
Therefore, since aa.oow is .58 and a d is 1.24, the obtained differ- 
ences are greater than are accounted for by chance alone. 
To determine the proportion of cases greater than those 

due to chance 1 we first find the ratio = ' ^ ==.47, and 

a# 1 .Jo 

we then enter Table 118 with this value to find that 35 
per cent of the 100 cases of differences between Paragraph 
Meaning and Computation scores are greater than can be 
accounted for by chance. 2 

TABLE 118 
PROPORTION OF DIFFERENCES IN EXCESS OF THE CHANCE PROPORTIONS 

Proportion of Proportion of Proportion of 

differences in differences in differences in 

excess of the excess of the excess of the 

chance chance chance 

proportion proportion proportion 

,02 95.0% .35 46.7% .70 17.1% 

.05 88.8% .40 41.5% .75 13.8% 

.10 79.8% .45 36.7% .80 10.8% 

.15 71.9% .50 32.3% .85 7.8% 

.20 64.7% .55 28.1% .90 5.1% 

.25 58.2% .60 24.2% .95 2.5% 

.30 52.2% .65 20.5% .99 0.5% 

This situation is represented in Figure 11 in which the 
shaded area represents the proportion of differences which 

iHereafter this percentage of cases showing differences greater than 
can be accounted for by chance will be spoken of as measurable disparity. 
2Kelley, T. L. Jour. Ed. Psych., September, 1923, p, 330. 



318 



TRAITS OF GIFTED CHILDREN 



are in excess of chance differences. The dotted curve repre- 
sents a distribution with o d .oo W as the standard deviation, and 
the full line curve represents a distribution with o d as the 
standard deviation. 

FIGURE 11 

SHOWING DISTRIBUTIONS OF FOUND AND TRUE 
DIFFERENCES BETWEEN SCORES 




The full line curve represents the distribution of 100 dif- 
ferences of the type z^-z 2 with the standard deviation of 1.24. 
The dotted line indicates a distribution with its standard 
deviation of 0.47 or a&oow for the differences between z scores 
for an individual. 

Using the inter correlations of Table 119 and the re- 



liability coefficients of Table 120, the ratio 



is cal- 



culated for each pair of the seven achievement tests, and for 
the four information tests. Certain ratios between some of 
the total scores are added for the purpose of comparing gen- 
eral reading and arithmetic ability with the other scores. 
In all cases <s d was larger than Cd.oow 9 hence every comparison 
shows measurable disparity; that is, some differences greater 
than can be expected from chance variation. Table 120 
presents the mean percentages of such differences. Arith- 
metic, Reading, and Information Totals are separated from 
the other percentages, as they obviously belong in another 
category. 

The most significant fact in this table is that the 100 
gifted children show measurable disparity for every pair 
of tests. For this group Music and Art Information and 
Language Usage show the least disparity. The mean of the 
percentages for the 55 comparisons involving the eleven 
tests is 29.3. Including the comparisons with Arithmetic 
Total, Reading Total, and Information Total adds 20 com- 



THE SPECIALIZATION OF ABILITIES 319 

TABLE 119 

INTERCORRELATIONS OF THE STANFORD ACHIEVEMENT TESTS FOR THE 
SPECIAL GROUP OF 100 GIFTED CHILDREN 



si a, stif! s i S |*5 

3l 31 II IlIlJIII Is' II I! I! s 



Arithmetic 
Reasoning .55 

Word 

Meaning .34 .37 

Sentence 

Meaning .30 .37 .75 

Paragraph 

Meaning .25 .36 .54 .47 

Language 

Usage .13 .10 .45 .45 .23 

Spelling .54 .45 .54 .46 .46 .35 

Science 

Information .29 .41 .47 .39 .24 .25 .41 

Language 

& Literature .24 .38 .64 .49 .41 .30 .38 .55 

History 

& Civics .44 .61 .62 .50 .39 .18 .54 .55 .66 

Music 

& Art .30 .36 .59 .47 .34 .40 .44 .51 .60 .54 

Reading 

Total .36 .43 .44 .57 .43 .60 .60 .54 .63 

Arithmetic 

Total .44 .42 .37 * .14 .63 .43 .38 .64 .41 .88 .46 

parisons and changes the mean to 30.4. Hence Figure 11 
with the shaded area showing 35 per cent involves only a 
little higher percentage than the type for this group. 

The average percentage of differences which are in ex- 
cess of the percentage due to chance for each of the 10 pairs 
having one test constant are shown in Table 121. 

From Table 121 it is evident that spelling ability as 
measured by the Stanford Achievement Test develops more 
independently of the other tests than any of the other abili- 
ties measured. Computation is a close second. The mean 
of all the percentages in Table 121 is 29. The five tests 
which are above this mean are measuring abilities which 
are largely independent of any measured by the other tests. 



320 TRAITS OF GIFTED CHILDREN 

TABLE 120 

THE PERCENTAGES OF DIFFERENCES IN EXCESS OF THE PERCENTAGES 
DUE TO CHANCE FACTORS 

fi 



II 11 II if ii II is I li ! 



Arith. 
Reas. 

Word 

Mean. 44% 40% 

Sent. 

Mean. 38% 33% 17% 

Par. 

Mean. 35% 29% 25% 23% 

Lang. 

Usage 28% 28% 19% 18% 21% 

Spell- 

ing 41% 41% 43% 36% 32% 24% 

Sci. 

Inf. 33% 28% 28% 27% 28% 20% 38% 

Lang. 

& Lit. 43% 38% 30% 30% 29% 23% 45% 23% 

Hist. 

& Civ. 36% 25% 29% 28% 28% 26% 37% 21% 24% 

Music 

& Art 35% 30% 21% 23% 26% 16% 34% 19% 27% 23% 

Read. 

Total 45% 39% 20% 44% 30% 33% 32% 27% 42% 

Arith. 

Total 44% 34% 32%/ 29% 39% 26% 41% 27% 32% 14% 42% 

Viewing Tables 120 and 121 as descriptive of the 
group of 100 gifted children, it is clear that they have fur- 
nished a substantial percentage of valid differences for 
every pair of tests. Furthermore there is a hierarchy of dif- 
ferences for the eleven tests. The greatest average per cent 
of differences occurs for the series of ten pairs of tests which 
have spelling in common, the next largest for the Compu- 
tation series, and the smallest percentage for the series 
which includes Language Usage as one member of each of 
the ten pairs. 

It is clear that in the group of 100 gifted children there 
is a significant inequality in the development of the abilities 



THE SPECIALIZATION OF ABILITIES 321 

TABLE 121 

THE MEAN PERCENTAGE OF MEASURABLE DISPARITY FOR EACH TEST 
WHEN IT Is COMPARED WITH ALL THE OTHER TESTS 

Percentage in 
Excess of Chance 
Constant Test Percentage 

Spelling 37% 

Computation 36% 

Reasoning 32% 

Language and Literature Information 31% 

Word Meaning 30% 

History and Civics Information 27.6% 

Paragraph Meaning 27,5% 

Sentence Meaning 27% 

Science Information 26% 

Music and Art Information 25% 

Language Usage 22% 

Mean 29% 

measured by the tests. Are these significant inequalities 
greater or less than those which would be encountered in an 
unselected or at least a less rigorously selected group of 
children ? 

KelleyV investigation of 96 pupils from four Palo Alto 
eighth grades furnishes data for such a comparison. 

Although the eighth grade children have been in school 
longer than our gifted group and are older chronologically, 
they probably do not average quite so high in mental age. 
The range of mental ages for the eighth graders is doubtless 
much greater than the 18 months' range of our gifted group. 
Just how these differences between the two groups would 
operate in their effect on the measurable disparity within 
individuals is difficult to say. Longer experience in school 
could be expected to bring up a child's ability in some of the 
subjects. For example, Arithmetic Reasoning, Spelling, and 
History and Civics Information, should be brought up to- 
ward the general level of an individual's ability by the rather 
uniform school training. On the other hand, given definite 
interests or stimuli for acquiring musical or literary infor- 
mation, the older eighth grade children have had several 
years in which to raise these functions above their general 
level. 

1 Kelley, T. L., op. cit. 



322 



TRAITS OF GIFTED CHILDREN 



Table 122 gives the data for 96 eighth grade children, 
together with the comparable data for the gifted. The upper 
figure in each cell is the percentage of differences in the test 
scores of individuals which is in excess of the chance dif- 
ferences for the gifted children, while the lower figure is the 
percentage for the eighth grade children. The mean percent- 
age of difference for the eighth grade children is 28.2; for the 
gifted children it is 30.7. These means indicate slightly more 
of unevenness for the gifted children. 

TABLE 122 

COMPARISON OF 100 GIFTED AND 96 CONTROL CHILDREN IN PERCENTAGE 

OF DIFFERENCES IN INDIVIDUAL TEST SCORES IN 

EXCESS OF THE CHANCE PERCENTAGE 



Arith. Arith. Word Sent Par. Lang. 
Comp. Reas. Mean. Mean. Mean. Usage 



Spell- Sci. 
ing Inf. 



Read. 
Total 



Arith. G 
Reas. C 


26% 










Word G 
Mean. C 


44% 
37% 


40% 
38% 








Sent. G 
Mean. C 


38% 
33% 


33% 
33% 


17% 
18% 






Par. G 
Mean. C 


35% 
28% 


29% 

27% 


25% 
22% 


23% 

17% 




Lang. G 
Usage C 


28% 
28% 


28% 
28% 


19% 
26% 


18% 
18% 


21% 

14% 


Spell- G 
ing C 


41% 

27% 


41% 

41% 


43% 

44% 


36% 

37% 


32% 
32% 


Sci. G 
Inf. C 


33% 
26% 


28% 
18% 


28% 
28% 


27% 
22% 


28% 
20% 


Read. G 
Total C 












Arith. G 
Total C 













24% 
29% 

20% 

24% 

20% 
20% 

29% 
29% 



38% 
33% 

44% 
44% 

39% 

37% 



30% 
26% 

29% 42% 
20% 36% 



There axe 35 pairs of traits compared in Table 122, 
of which 13 pairs show variation of 1 per cent or less. If we 
do not count the 7 pairs which include the total scores we 
find the gifted children show greater unevenness in 15, or 
54 per cent, of the compared pairs, and that the gifted and 
control groups have equal measurable disparity in 7, or 25 
per cent of the comparisons. The children of the eighth 



THE SPECIALIZATION OF ABILITIES 323 

grade control group show greater measurable disparity in 
only 6, or 21.4 per cent of the pairs. If these differences be- 
tween test scores are definitely related to intrinsic differ- 
ences in traits, this is evidence that the gifted children are 
slightly more specialized and even more truly "real persons 
with specific and unique mental mechanisms" than are un- 
selected children. 

One reason for thinking the differences found may not 
be in perfect correlation with the true differences is the 
fact that these tests are not all equally reliable. We may 
estimate what the differences would be if all the tests were 
equally reliable., The mean reliability coefficient is .829 if 
all tests are weighted equally and the totals of Arithmetic, 
Reading, and Information are omitted. 

The correlation between Arithmetic Reasoning and 
Arithmetic Computation is .551, but if the two tests were 
perfectly reliable this correlation would be 

_ r l2 .551 

r w= V^ V^~ V-898 V-864 

This r oow is the probable correlation between the true 
scores of the two traits. The formula is Spearman's formula 
for correction for attenuation. 

But let us suppose that the two tests are equally reliable 
and that their reliability is .829, or the same as the mean 
reliability for all the tests. Then substituting their value for 
jTu and r 2H and solving for r l2 we have 

r 12 = .626 V.829" V^829"= (.626) (.829) = .519. 
Next we can find the ratio -^^ for Arithmetic Computa- 

Cfd 

tion and Arithmetic Reasoning under the assumed condition 
that their reliabilities are both .829. 



^^ = ' ' = .597. Using this ratio to enter Table 
<fe V2-2(.519) 

119 we find .24, which is the measurable disparity be- 
tween Computation and Arithmetic Reasoning if both tests 
have the reliability of .829. Table 123 is built up in this 
way. It shows the best estimate of what the relationships 
between the abilities of the children in this group would be 
if the reliabilities were all equal and all .829. 



324 TRAITS OF GIFTED CHILDREN 

TABLE 123 

PERCENTAGE OF DIFFERENCES IN INDIVIDUAL TEST SCORES IN EXCESS OF 

THE CHANCE PERCENTAGE IN CASE THE RELIABILITY OF 

EACH TEST Is .829 (GIFTED CHILDREN) 


J * 

II 11 A II II li is 11 1 Ii 

Arith. 
Reas. 24% 

Word 

Mean. 32% 30% 

Sent. 

Mean. 31% 31% 13% 

Par. 

Mean. 34% 31% 24% 26% 

Lang. 

Usage 37% 38% 25% 24% 33% 

Spell- 
ing 26% 29% 26% 28% 27% 30% 

Sci. 

Inf. 33% 29% 23% 29% 34% 32% 30% 

Lang. 

&Lit. 35% 31% 20% 26% 28% 31% 31% 23% 

Hist. 

&Civ, 29% 21% 21% 26% 30% 35% 26% 22% 18% 

Music 

& Art 32% 31% 21% 26% 31% 26% 29% 23% 20% 23% 

The results of Kelley's application of this method to the 
eighth grade pupils, together with the comparable results 
from Tahle 123, are shown in Table 124. 

Table 124 supplies another comparison of the two 
groups of children. The mean of all the per cents in Table 
124 is 28.89 for the gifted group and 27.82 for the control 
group. In the 28 cells of this table we find the gifted children 
showing the larger percentage 13 times, the eighth grade 
children 12 times, and the two an equal per cent 3 times. 

Comparing these findings with those recorded in Table 
122, it appears that under the condition of uniform relia- 



THE SPECIALIZATION OF ABILITIES 



325 



TABLE 124 

PERCENTAGES OF DIFFERENCES IN INDIVIDUAL TEST SCORES IN EXCESS OF 

THE CHANCE PERCENTAGE IN CASE THE RELIABILITY Is 

THE MEAN FOR THE GROUP 

(Control group .824. Gifted group .829) 



Arith. 
Reas. 

Word 
Mean. 


G 
C 

G 
C 


Arith. 
Comp. 

24% 
34% 

32% 

39% 


Arith. 
Reas. 

30% 

30% 


Word 
Mean. 


Sent. 
Mean. 


Par. 
Mean. 


Lang. 
Usage 


Spell- 
ing 


Sent. 
Mean. 


G 
C 


31% 

40% 


31% 
26% 


13% 
13% 










Par. 
Mean. 


G 
C 


34% 
35% 


31% 
26% 


24% 

14% 


26% 

17% 








Lang. 
Usage 


G 
C 


37% 
39% 


38% 
33% 


25% 
24% 


24% 
23% 


33% 

17% 






Spell- 
ing 


G 
C 


26% 
31% 


29% 
36% 


26% 
28% 


28% 
33% 


27% 
25% 


30% 
30% 




Sci. 
Inf. 


G 
C 


33% 

35% 


29% 
20% 


23% 
24% 


29% 
25% 


34% 
21% 


32% 

30% 


30% 

31% 



bility the gifted and control groups show less difference in 
the amounts of unevenness than they show as tested by the 
tests used. In Table 122 ten cells out of 28 show a difference 
of 1 or less, while in Table 124 eight cells show similar 
differences. 

Table 125 shows the mean percentage of measurable 
disparity in each test when it is compared with all other 
tests. 

The data of Table 125 show many interesting facts 
concerning the parallelism of development of the traits 
measured. Those which are most nearly parallel in their 
development as revealed in the tests of the gifted children, 
and their percentages of measurable disparity, are as 
follows : 

Music and Art vs. Science Information 16% 
Word Meaning vs. Sentence Meaning 17% 

Language Usage vs. Sentence Meaning 18% 
Language Usage vs. Word Meaning 19% 

Music and Art vs. Science Information 19% 
Language Usage vs. Science Information 20% 



326 



TRAITS OF GIFTED CHILDREN 



TABLE 125 

THE MEAN PERCENTAGE OF MEASURABLE DISPARITY FOR EACH TEST 
WHEN IT Is COMPARED WITH ALL THE OTHER TESTS 

Rank 



Reliability 
.829 



From Table Rank from Mean 
121 Table 121 Rank 



Arithmetic 
Computation 


31.3% 


Language 
Usage 


31.1% 


Paragraph 
Meaning 


29.8% 


Arithmetic 
Reasoning 


29.5% 


Spelling 


28.2% 


Science 
Information 


27.8% 


Language and 
Literature 
Information 


26.3% 


Music and Art 
Information 


26.2% 


Sentence 
Meaning 


26.0% 


History and 
Civics 
Information 


25.1% 


Word 
Meaning 


23.5% 



7 
8 
9 

10 
11 



36 
22 
27.5 

32 
37 

26 

31 
25 
27 

27.6 
30 



2 

11 

7 

3 
1 



4 

10 
8 

6 
5 



7.5 

5.5 

9 

8.5 

8 
8 



Showing most independence of development, or least par- 
allelism in its relation to all other traits measured, is Spell- 
ing. In the case of Word Meaning and Spelling the dispar- 
ity is 42.5, which raises the question whether the meaning 
of words and the spelling of words are not less intimately 
related than the majority of school subjects. When we find 
the mean of the percentages of measurable disparity be- 
tween each test and all the others, Spelling has the largest 
percentage and Language Usage the smallest. This may be 
seen in Table 125, where these facts are given in the 
third column under the caption "From Table 121." From 
this table it is evident that the condition of uniform reliabil- 
ity considerably alters the rank order of the tests in their 
measurable disparity. 



THE SPECIALIZATION OF ABILITIES 327 

The condition of uniform reliability also affects the mean 
of the differences. Taking the reliabilities as found, the 
differences in Table 122 were greater for the gifted chil- 
dren in 54 per cent of cases, greater for the control group in 
21 per cent of cases, and equal in 25 per cent. Under the con- 
dition of uniform reliability the paired comparisons of Table 
125 show greater measurable disparity as follows: gifted, 
46.4; control, 42.8; equal disparity, 10.7. 

Although these results so far indicate a slightly greater 
unevenness for the gifted children, they must not be allowed 
to obscure the more important fact that both groups of chil- 
dren are characterized by considerable unevenness of their 
abilities as measured by these tests. 

A FURTHER COMPARISON BY MEANS OF TYPE GROUPS OF GIFTED 
AND UNSELECTED CHILDREN 

The comparison just completed is based upon the relia- 
bility coefficients and the intercorrelations between the 
tests. A more direct and more nearly one to one comparison 
is desirable. Furthermore, the first comparison demanded 
that two groups be chosen, and it may be that other groups 
would show other results. The method used in the following 
study furnishes a check on the earlier method and at the 
same time throws light on the popular view as to the sup- 
posed compensatory distribution of abilities. 

The problem of the specialization of abilities presents it- 
self to the teacher and the parent in a way which is sug- 
gested by the descriptive terms, "good readers" and "poor 
readers," "good spellers" and "poor spellers," "good calcu- 
lators." Let us assume that goodness and poorness in these 
terms are relative within one individual's equipment. Then 
a "good reader'* 1 would be a pupil who could read better 
than he could spell, write, or calculate, regardless of his gen- 
eral mental level. If we limit our discussion to children 
showing such specialization we may then ask, "Are the 'good 
readers' among the gifted children more or less specialized 
than the 'good readers' among other children?" Popular 
opinion seems to hold that such specialization occurs more 
frequently among supernormal and subnormal children. 

Hereafter "good reader," "good calculator," etc., are used in the sense 
described above and hence do not refer to the relative "goodness" of these 
abilities as compared with the abilities of other pupils. 



328 TRAITS OF GIFTED CHILDREN 

To make this comparison three groups of "good readers" 
were selected. First for a control group we have the 307 rep- 
resentative 12 year old children for whom we have scores 
from the Stanford Achievement Tests. 1 

The criterion of selection of "good readers" was the dif- 
ference between the Standard Scores for the Reading Total 
and the Arithmetic Total. The difference between these 
scores was calculated for each of the 307 cases, and then the 
20 showing the reading scores most in excess of their arith- 
metic scores were arbitrarily chosen as "good readers." Per- 
haps these were as much poor calculators as good readers 
but as the same method is used for selecting the "good read- 
ers" from the gifted children, it furnishes a valid basis for 
comparison. Table 126 shows the standard scores for 
the 20 "good readers" from the 307 representative 12 year 
old children. The mean grade location for these twenty 12 
year old children is 6.77 grades, which is the equivalent of 
about the eighth month in the sixth grade, or May in the 
school year, if there were no high and low divisions. 

Two groups of gifted children are compared with these 
20 "good readers" selected from typical 12 year old children: 
(1) Twenty 12 year old gifted "good readers" whose stand- 
ard scores are shown in Table 127; and (2) 20 sixth grade 
"good readers" whose standard scores are shown in Table 
128. These gifted "good readers" were selected by the same 
method as that followed in the selection of the typical 12 
year old children. 

The 12 year old gifted children were selected because 
they have in common with the typical 12 year old children 
12 years of experience. The gifted children have had a 
more extensive school training, since their grade place- 
ment ranges from the seventh to the ninth, with the mean 
in the low eighth, while the typical 12 year old "good read- 
ers" have a mean grade placement of high sixth and a range 
from high fourth to high eighth. This additional schooling, 
which averages a little more than a grade to each child, 
would seem to give the gifted children an opportunity for 
more varied development and hence greater specialization. 

*For these data the writer is indebted to Professor Truman L. Kelley. 
These 12 year olds were selected from all of the grades from the 2nd 
through the 9th. They are a very representative sampling of 12 year old 
children. 



THE SPECIALIZATION OF ABILITIES 329 



f f 



uonBuuoinT ocMWi>oi00oococooieocviooooTHCcii-icooi> TH 

a^TTQTOo THOOO^I>I>05C^C^OI>I>OOOCOTH^O OO 
f rl TH ^ " ' ^ TH TH TH v4 r4 * " tH " f f " 



p 

J 



TH ,H | | 

8irprB9jv; 

90U81UOS ^ 



Til 






CO 

o 



f f " f * " f f 



O mTTaO 

g '"Ht^^S c^05OI>C50OOqi>;iOCSJOI>C3CaC3xCl>|>Oa 

C/D " " TH " TH TH CSJ TH " * TH TH " I* f TH f I* f TH 



f f I* f 



. . . fl 

. a u q pj w fa o u co -< Q co PE; o sd . w . ^ S 



330 TRAITS OF GIFTED CHILDREN 

TABLE 127 



STANDARD SCORES 


OF 20 GIFTED SIXTH GRADE 

A 1 . .. . 


"GOOD READERS" 

s 


"3 
' 6 


1 


If 


fcc to 

i s 


Is 


50 
33 


ll 


ll 


C2 


ri 


|H M 

co 5 


"S3 

O 


11 


40 


B 


10.9 


2.38 


2.26 


.62 


.64 


2.77 


3.25 


2.28 


2.55 


2.72 


1.80 


47 


G 


10.5 


3.61 


2.85 


1.48 


1.99 


2.53 


2.71 


1.86 


2.48 


2.32 


1.76 


56 


B 


10.3 


.62 


1.97 


.12 


1.37 


1.58 


1.49 


1.44 


2.95 


1.52 


.48 


84 


G 


9.8 


4.84 


3.73 


3.34 


2.48 


3.95 


3.16 


3.26 


4.59 


3.43 


2.96 


116 


B 


8.7 


4.00 


3.20 


2.37 


5.11 


4.26 


4.36 


3.68 


4.59 


3.99 


3.40 


134 


G 


11.2 


1.88 


1.70 


1.21 


1.17 


2.32 


1.81 


1.04 


2.08 


1.70 


1.18 


139 


B 


10.4 


2.55 


2.73 


1.60 


3.02 


2.89 


3.31 


1.69 


4.18 


2.88 


2.24 


195 


B 


11.3 


1.88 


1.55 


.50 


1.87 


1.45 


2.30 


1.44 


2.73 


1.76 


1.10 


201 


B 


12.2 


1.89 


1.11 


1.29 


1.22 


2.31 


1.79 


2.09 


2.39 


2.06 


1.27 


207 


B 


12.1 


1.32 


1.92 


.62 


1.53 


1.83 


1.23 


1.43 


.97 


1.49 


1.05 


272 


G 


11.4 


1.28 


1.81 


1.07 


1.17 


2.32 


2.14 


1.20 


2.53 


1.87 


1.10 


274 


B 


11.5 


1.50 


2.06 


1.21 


2.05 


2.32 


2.19 


2.00 


3.28 


2.22 


1.57 


279 


B 


11.2 


1.43 


1.86 


.50 


1.87 


1.34 


1.87 


1.12 


2.18 


1.46 


1.10 


294 


G 


9.8 


1.86 


2.88 


2.21 


2.25 


2.31 


3.81 


2.40 


1.53 


2.74 


2.21 


309 


B 


11.6 


2.25 


2.57 


.07 


2.05 


1.01 


1.65 


1.52 


2.13 


1.46 


.94 


370 


G 


11.4 


2.33 


1.19 


.36 


.47 


1.78 


2.36 


1.84 


2.13 


2.04 


.39 


372 


B 


10.8 


3.08 


2.50 


1.11 


2.61 


2.53 


2.50 


2.11 


1.87 


2.38 


1.77 


465 


B 


11.3 


1.28 


2.62 


1.36 


1.17 


1.88 


1.65 


1.60 


1.48 


1.76 


1.25 


511 


G 


10.6 


2.38 


2.14 


1.60 


1.37 


2.77 


2.30 


1.44 


2.41 


2.10 


1.60 


521 


G 


10.3 


2.03 


2.03 


.86 


2.61 


2.29 


3.58 


1.94 


2.07 


2.52 


1.60 


Mean 




10.86 


2.22 


2.23 


1.18 


2.05 


2.32 


2.47 


1.82 


2.46 


2.22 


1.54 



TABLE 128 
STANDARD SCORES OF 20 GIFTED 12 YEAR OLD "GOOD READERS' 



3 S? 5 a a $2 * S S S S . . "2 "3 r 



-2 * U G% ^ S s S3 g S S S S .. g gg 

feo S ** B W & o o H > iS <3 tS rt^'Sd v ^ 

00 Z CO O i-IP t/2 U <} PH f> S CO <i P-i K5 CO HH PH H <{ H 

120 G 8.7 2.46 1.36 1.43 1.22 2.25 2.01 1.81 1.68 2.00 1.34 

128 B 8.3 2.78 2.06 .05 2.31 2.55 1.79 1.62 2.10 1.95 1.05 

135 G 7.3 2.46 1.36 1.02 1.53 2.37 1.68 1.53 2.44 1.83 1.27 

137 G 8.8 1.48 1.26 .62 1.22 2.13 1.40 2.28 1.59 1.95 .91 

140 G 7.8 1.97 .96 .05 .59 1.47 1.40 1.15 1.87 1.32 .25 

253 B 7.7 2.46 1.87 .49 1.68 2.31 1.90 .68 2.29 1.56 1.05 

276 B 7.8 1,07 1.92 1.02 '1.68 2.25 1.90 1.34 1.40 1.79 1.34 

297 G 7.7 1.80 2.12 .75 1.06 2.01 2.12 1.62 1.92 1.90 .91 

302 G 8.2 2.62 2.22 1.16 1.53 2.37 2.12 1.81 1.63 2.08 1.34 

307 G 7.7 2.13 1.46 .89 2.00 2.07 1.79 1.81 2.06 1.88 1.41 

320 B 7.8 2.05 1.82 1.16 1.53 1.77 1.79 1.90 1.40 1,84 1.34 

326 G 8.7 2,62 1.82 .89 .90 2.31 1.84 2.19 2.10 2.11 .91 

368 B 8.3 2.13 2.43 1.29 .75 2.37 2.01 1.15 1.87 1.79 1.05 

385 G 7.2 2.62 1.72 .75 1.84 2.07 1.90 1.43 1.73 1.77 1.27 

406 G 8.8 2.13 1.77 .89 2.00 2.07 1.68 2.28 2.15 2.02 1.41 

427 G 8.2 2.13 1.77 1.29 1.53 2.49 2.12 2.56 1.92 2.41 1.41 

472 G 7.7 1.64 1.31 .49 1.06 2.07 2.18 2.28 1.40 2.18 .76 

490 B 7.3 1.64 1.82 .62 2.31 1.95 2.12 1.62 1.59 1.88 1.41 

494 G 9.3 2.78 2.22 .89 1.84 2.43 2.12 3.04 1.63 2.06 1.34 

545 B 8.7 2.78 2.07 1.02 1.84 2.25 2.01 2.09 2.20 2.11 1.41 

Mean 8.1 2.18 1.76 .84 1.52 2.18 1.89 1.81 1.85 1.90 1.15 



THE SPECIALIZATION OF ABILITIES 331 

This expectation is not fulfilled, as we shall see when we ex- 
amine the data of Table 129. 

The sixth grade gifted "good readers" (Table 128) 
were selected for comparison with the typical 12 year old 
"good readers" because the latter have a mean grade place- 
ment of 6.77, or high sixth grade. Then sixth grade gifted 
"good readers" probably have a school experience and train- 
ing quite similar to that of the typical 12 year old "good 
readers." 

From these three tables (126, 127, and 128) we may 
secure the numerical equivalence of a composite photo- 
graph of the profile charts of each group of 20 "good 
readers." This composite photograph for each group is 
described by the mean standard scores given at the bottom 
of its table. These mean scores assembled in one table 
(Table 129) give an opportunity for a direct comparison 
of the three groups. 

TABLE 129 
NUMERICAL PROFILES OF THREE GROUPS OF "GOOD READERS" 

Good Lang. Spell- Arith. Word Sent. Par. Sci. Read. Arith. 

Readers Usage ing Comp. Reas. Mean. Mean. Mean. Inf. Total Total 

12yr. old 

gifted 2.18 1.76 .84 1.52 2.18 1.89 1.81 1.85 1.90 1.15 

Typical 

12 yr. olds .71 .45 -.03 -.02 .98 1.15 .79 .84 .98 .10 

6th grade 

gifted 2.22 2.23 1.18 2.05 2.32 2.47 1.82 2.46 2.22 1.54 

From Table 129 and Figure 12 it is apparent that the 
gifted "good readers" are superior to the typical 12 year old 
children in all the abilities measured. 

An inspection of the three composite profiles of Figure 
12 does not reveal any marked differences in unevenness. 
But there is a defect of such graphical representation. Ob- 
viously, in this picture the emphasis is on the differences 
between the standard scores for adjacent ordinates, such as 
Language Usage and Spelling, while the differences between 
non-contiguous ordinates such as Language Usage and 
Science Information is almost lost to view. To bring out 
these concealed differences Tables 130 and 131 have been 
prepared. In these tables the differences between the mean 



332 TRAITS OF GIFTED CHILDREN 

FIGURE 12 
PROFILES OF THREE GROUPS OF "GOOD READERS" 





i 

J 




U j 

" 


: <t 

\ & 


: 1 


! J 


ft 

' 1 


f eg 

!*- 


^ 


^. 





i 


> 

2 
5 ff] 


- 

^ < 


\ 
*f 




! t 

> c 
: v. 


- 2 
i J 


* 
CO 


j 


I 


4 






















35- 






















3 
as- 

V) 




\ 






/> 


s' 

\ 


\ 


/ 

-Jl . 


s s 


\~ 


o 




\ 


\ 


/ 


i 


\ 


^-^ 


7 


*** 


\\ 


7T i5- 






\\ 


/ ^ 












\ ^ 

% 


t 






\\ 
















73 

r . 






\ 1 


A 




^ 










s. 

(0 

0-5- 




^ 


\ 

\ 




i 




\ 


^ 


-^^^ 


K 


-05- 






















-1 








w Gifte 


d sixth 


grade 


"good i 


eaders. J 


> 












... Gifte 


d 12 ye 


;ar old 


"good i 


eaders.' 


' 





THE SPECIALIZATION OF ABILITIES 333 

standard scores are expressed in terms of standard scores. 
They could be expressed in terms of the standard errors or 
the probable errors of such differences (o d ., x>w or P.E. &,,), 
but such standard errors would be the same for all the 
scores of 12 year old children and but slightly different for 
9, 10, and 11 year old children of the sixth grade group and 
hence would not materially change our comparisons. 

TABLE 130 

DIFFERENCES BETWEEN MEAN STANDARD SCORES OF 20 TYPICAL 12 YEAR 
OLD "GOOD READERS" AND 20 GIFTED 12 YEAR OLD "Gooo READERS" 

Lang. Spell- Compu- Arith. Word Sent Par. Sci. 
ing tation Reas. Mean. Mean. Mean. Inf. 



Spell- 
ing 

Compu- 
tation 

Arithmetic 
Reasoning 

Word 


G 
C 

G 
C 

G 
C 

G 


.42 
.26 

1.34 
.74 

.66 
.73 

.00 


.92 
.48 

.24 
.47 

.42 


.68 
.01 

1.34 


.66 










Meaning 


C 


.27 


.53 


1.01 


1.00 










Sentence 


G 


.29 


.13 


1.05 


.37 


.29 








Meaning 


C 


.44 


.70 


1.18 


1.17 


.17 








Paragraph 


G 


.37 


.05 


.97 


.29 


.37 


.08 






Meaning 


G 


.08 


.34 


.82 


.81 


.19 


,36 






Science 


G 


.33 


.09 


1.01 


.33 


.33 


.04 


.04 




Information 


C 


.13 


.39 


.87 


.86 


.14 


.31 


.05 




Mean 


G 


.49 


.32 


1.04 


.46 


.49 


.32 


.31 


.31 


Mean 


C 


.38 


.45 


.73 


.72 


.47 


.62 


.38 


.39 



The standard errors (od.< w ) for 12 year old children 
never rise above 60 and are usually below 50, hence we may 
think of the differences expressed in our present study as 
multiplied by 2 if we wish to think of their probable va- 
lidity. 

In Table 130 the differences between the mean scores 
in Language Usage and Spelling are .42 standard deviation 
for the gifted and .26 for the control group. Hence the differ- 
ence between these two scores is greater in the case of the 
gifted group. There are 28 such comparisons in this table 
exclusive of those involving the Arithmetic Total and Read- 



334 



TRAITS OF GIFTED CHILDREN 



TABLE 131 

DIFFERENCES BETWEEN THE MEAN STANDARD SCORES OF 20 TYPICAL 

12 YEAR OLD "Gooo READERS" AND 20 GIFTED 

SIXTH GRADE "Gooo READERS" 

Lang. Spell- Compu- Arith. Word Sent. Par. Sci. 
Usage ing tation Reas. Mean. Mean. Mean. Inf. 



Spell- 
ing 

Compu- 
tation 

Arithmetic 


G 
C 

G 
C 

G 


.01 
.26 

1.04 
.74 

.17 


1.05 
.48 

.18 


.87 










Reasoning 


C 


.73 


.47 


.01 










Word 


G 


.10 


.09 


1.14 


.27 








Meaning 


C 


.27 


.53 


1.01 


1.00 








Sentence 


G 


.25 


.24 


1.29 


.42 


.15 






Meaning 


C 


.44 


.70 


1.18 


1.17 


.17 






Paragraph 


G 


.40 


.41 


.64 


.23 


.50 


.65 




Meaning 


C 


.08 


.34 


.82 


.81 


.19 


.36 




Science 


G 


.24 


.23 


1.28 


.41 


.14 


.01 


.64 


Information 


C 


.13 


.39 


.87 


.86 


.14 


.31 


.05 



Mean G .33 .33 1.04 .36 .34 .43 .50 

Mean C .38 .45 .73 .72 .47 .62 .38 



.42 
.39 



ing Total scores by which the groups were selected. In these 
28 comparisons the differences are greater for the control 
group 16 times, and greater for the gifted group 12 times. At 
the bottom of this table are displayed the means of the dif- 
ferences between each test and all the other tests. Thus the 
mean of all the differences between Language Usage and the 
other test scores for the gifted group is .49, and for the con- 
trol group it is .38. Of such means there are 8 if we again 
exclude all differences which involve total scores in Arith- 
metic and Reading. Of these 8 differences 5 are greater for 
the control group and 3 are greater for the gifted group. If 
we add all the 28 differences for the control group the total 
is 14.51, while for the gifted group it is 13.11. Therefore by 
every comparison, the composite profiles of gifted 12 year 
old "good readers" show slightly less of unevenness or spe- 
cialization than does the composite profile of typical 12 year 
old "good readers. 95 



THE SPECIALIZATION OF ABILITIES 335 

A similar analysis of Table 131 indicates that the com- 
posite profile of the 20 gifted sixth grade "good readers" also 
shows slightly less of unevenness than does the composite 
profile of the 20 typical 12 year old "good readers," The com- 
parisons used in Tables 130 and 131 are summarized below. 

Total Differences Means for 
Differences Greater each test 



Gifted 12 year olds 
Typical 12 " " 


13.11 
14.51 


12 
16 


3 
5 


Gifted sixth graders 
Typical 12 year olds 


13.05 
14.51 


12 
15 


3 
5 



"Good readers" if they represent a type, must be one of 
many types. Since they were selected with reference to their 
total scores in Arithmetic, it is interesting to examine their 
complementary type, which we shall call "good calculators." 
Here again our term may be a misnomer, for these may be 
poor readers as well as good calculators, but they are good 
in arithmetic when their arithmetic scores are compared 
with their reading scores. Again three groups were selected, 
typical 12 year olds, gifted 12 year olds, and gifted sixth 
graders. 

The composite profiles of these three kinds of "good 
calculators" are given in Figure 13. 

As in the case of the "good readers" the profiles of the 
gifted children who are "good calculators" show a tendency 
for Arithmetic Reasoning to be higher than Computation. 

When typical 12 year old "good calculators" are com- 
pared with the gifted 12 year old "good calculators," it de- 
velops that there is more of unevenness in the control group. 
Of the 28 paired comparisons, one pair shows the same dif- 
ference for the two groups, 4 show greater differences for 
the gifted, and 23 show greater differences for the typical 
12 year old children. The total of the 28 differences for the 
control group is 14.82, while the total for the gifted group is 
6.99. All of the mean differences for one test compared with 
all others are greater for the control group. 

How do the gifted sixth grade "good calculators" com- 
pare with unselected 12 year old "good calculators" in 
specialization of abilities? This comparison reverses the one 



336 



TRAITS OF GIFTED CHILDREN 



35- 



2,5- 



(0 


O 



1.5- 



0.5- 



-05- 



FIGURE 13 
PROFILES OF THREE GROUPS OF "Gooo CALCULATORS 






M <{> 

c5 2: 

-C T3 

t 1 



< J 

5 S 



3 



\ 



V< 






\ 
\ 







Gifted sixth grade "good calculators." 

Gifted 12 year old "good calculators." 
Typical 12 year old "good calculators." 



THE SPECIALIZATION OF ABILITIES 337 

just given. Here we find but 12 of the 28 pairs of differences 
to be greater for the control group and 16 greater for the 
gifted sixth graders. The total of the 28 differences for the 
gifted sixth graders is 17.63, as compared with the total 
for the control group of 14.82. The means of each test against 
all others are greater for the gifted in 7 cases and greater for 
the control group in one. All of the evidence here is indica- 
tive of greater unevenness for gifted sixth grade "good cal- 
culators" than for unselected 12 year old "good calculators." 

The selection of "good readers" and "good calculators" 
is based upon total scores in Reading and in Arithmetic, re- 
spectively. In each selection the differences considered are 
consequently differences between total scores, which in their 
turn are based upon tests having considerable independence. 
Because of this use of total scores, it seemed advisable to 
use another selection based upon a comparison with the 
Spelling scores. Three groups were chosen, because in their 
respective larger groups they showed the greatest differ- 
ences between the score in Spelling and that in Reading 
Total. We have called these groups "good spellers." The 
groups from which these "good spellers" were taken are the 
same that were used before: typical 12 year old children, 
gifted 12 year old children, and gifted sixth grade children. 

The composite profiles of these three selections of scores 
are found in Figure 14. The three profiles are nearly parallel. 
A summary of the differences adds to this graphic testimony 
that the three groups have similar unevenness. Comparing 
the typical group with the gifted 12 year old children reveals 
15 of the 28 differences to be greater for the gifted and 13 
greater for the typical children. Of the mean differences 
when one test is compared with all others, 5 are greater for 
the control group and 3 for the gifted. If there is any differ- 
ence between these groups it is in favor of slightly greater 
unevenness for the control group, but the difference is neg- 
ligible. 

The gifted sixth grade "good spellers" are more uneven 
than the typical 12 year old "good spellers." This result is 
more definite than that of the preceding comparison. In the 
28 pairs the gifted show greater differences 17 times, and of 
the 8 means when one test is compared with all others 7 are 
larger for the gifted group. 



338 



TRAITS OF GIFTED CHILDREN 



FIGURE 14 
PROFILES OF THREE GROUPS OF "Gooo SPELLERS'* 









1 






"S 



35- 



Z.5- 



U 





CO 



0.5- - 



-0.5- 






\x 



x\ 









s\ 














. Gifted sixth grade "good spellers." 
. Gifted 12 year old "good spellers." 
, Typical 12 year old "good spellers." 



THE SPECIALIZATION OF ABILITIES 339 

CONCLUSION FROM THE Two METHODS OF COMPARISON 

We can now bring together the results of the comparison 
by means of Kelley's ratio and the comparison of the groups 
of 20 each by means of composite profiles. 

The first method shows the gifted to be slightly more un- 
even, although the difference in unevenness is not so signifi- 
cant as the resemblance. In other words, the fact that both 
the eighth graders and the gifted children show significant 
unevenness is the most important conclusion. In comparing 
the first n\e|hod with the second (the threefold selection), 
it must be recalled that the latter method deals with the 
comparison of the differences between obtained standard 
scores. It is probable that the first method is to be relied 
upon as giving the more trustworthy estimate, since it deals 
with the best estimates of the true scores. The study of the 
composite profiles attempts to select types and compares 
gifted and control children of approximately the same type. 

It is conceivable that the slightly greater unevenness of 
the group of gifted children with MA. 14-15.4, as compared 
with eighth grade children, is due to the criterion of selec- 
tion. On the other hand, the threefold comparisons show 
greater unevenness for the gifted children in two out of the 
six comparisons made. It is possible that a selection of other 
types on the same principle would show an equal amount of 
unevenness for the gifted and control groups, and it is even 
possible that gifted groups might show greater unevenness 
in a majority of the types selected. The lack of conspicuous 
differences between gifted and unselected children in the 
specialization of the abilities is the outstanding fact. Gifted 
children are unique individuals, but unselected children are 
no less so. Gifted children may be more successful special- 
ists and thus attract more attention, but it should be ob- 
served that they are superior in all abilities. On the other 
hand, children of ordinary or mediocre general intelligence 
often possess specialized abilities that are overlooked be- 
cause they are not so conspicuous as when they appear in 
the gifted specialist. Children on all levels of mental ability, 
must be recognized as unique individuals with unique edu- 
cational and vocational needs. 



340 TRAITS OF GIFTED CHILDREN 

NATURE OF THE UNEVENNESS IN THE ABILITIES OF 
GIFTED CHILDREN 

The conclusion that neither superior nor ordinary in- 
telligence is predictive of types of unevenness, leaves us the 
task of the description of the unevenness of individual chil- 
dren. When this is accomplished for our gifted group there 
are numerous possibilities of studying these descriptions in 
various combinations. 

It will be recalled that the 643 gifted children were tested 
by means of 12 tests, the 7 tests of the Stanford Achieve- 
ment battery, 4 tests from an information battery, and the 
Stanford-Binet The scores from these tests are used in the 
following study. As age groups below 6 and above 13 were 
represented by few cases only, the scores of children be- 
tween and including these ages were used. A few incom- 
plete scores made it necessary to reduce the group to about 
550, the number varying slightly in various portions of the 
treatment. 

The composite profiles used in the preceding section rep- 
resent the graphic method of description of the surface of 
abilities. A similar profile was prepared for each child 
studied. Figure 15 illustrates this type of profile chart. 

Figure 15 is the profile for A. K. The heavy line at "0" 
standard score represents the average scores made in the 
various tests by 10 year old children. (These average scores 
are given in Table 133, page 344.) The lines running parallel 
to the line of the mean scores mark off standard scores above 
and below the means. Thus the standard score on the or- 
dinate for Mental Age is a little less than 4, indicating that 
A. K's mental age is nearly 4 standard deviations above the 
average M.A. for 10 year old children. 

The scores from tests of Paragraph Meaning, Word 
Meaning, and Sentence Meaning were added to make up the 
score for Reading Total; the scores for Computation and 
Arithmetic Reasoning were added to make up the Arith- 
metic Total. Similarly the scores on the information tests 
were combined to make the Information Total. Hence the 
scores for Reading Total, Arithmetic Total, and Information 
Total are separated by a space to indicate that they are in a 
different category from the other scores. The solid line 
which crosses the page in a very irregular course connects 



THE SPECIALIZATION OF ABILITIES 



341 



the points which represent A. K's scores on the tests indi- 
cated. This line is A. K's profile of abilities in so far as we 
have measured them. It starts at the left from a point on the 
perpendicular which represents a Stanford-Binet M.A. of 
14-10, which is the equivalent of a standard score of 3.8. This 



FIGURE 15 
PROFILE OF A. K. 

(Boy; age, 10-3; mental age, 14-10; IQ, 144; grade, high 4.) 



1 

4 

^ 

s 

7 

8 


o 5 
CO 

^3 

C 

1 



A 


V) Q C 9 C o A -~ - 
XO u O(0 <0 flj J t <p < J " 

Q0r^>^: tt> <D S **S^-2l;* 
53.5^r|s^ - - :^ } 2i2|J 

! f I ! 1 I I ! i } I 1 1 1 i 








































































































/ 


s 


























/ 


\ 


5 
4 
3 






\ 
















/ 








1 


\ 


\'' 


\ 


V 






*>> 


/s 


*-~~ 


*" 


*"*"* 




'7 




\ 


S 

\ 


'' / 


^ 














\ 


7 


A<rej 


70TI77} 


\ 


/ 

veof 


3 

















































Solid line indicates obtained standard scores. 
Broken line indicates age equivalents for mental age. 

line finds its lowest point at Computation. On the other hand, 
A. K's score in History and Civics Information is more than 
5 standard deviations above the mean. All of his scores 
in the information tests are well above their means. Thus 
the distances on the perpendicular lines between the line of 
the means and the profile line are the measure of the dif- 



342 TRAITS OF GIFTED CHILDREN 

ferences between A. K's accomplishment and the average 
accomplishment of 10 year old children. 

Some might wish to compare a pupil's accomplishment 
with the average accomplishment of children of the same 
intelligence level. This comparison has been attempted by 
means of the accomplishment quotient. Such a comparison is 
found in Figure 15. The dotted line on A. K's profile crosses 
points which are the age equivalents in scores on the several 
tests corresponding to M.A. 14-10. That is, these age equiva- 
lents are estimated to be the average scores of children 14 
years, 10 months old. It must be remembered, however, that 
the units of comparison along the perpendicular lines are not 
the standard deviations for the distributions of scores made 
by 14 year old children, but are still the standard deviations 
for 10 year old children. We know that it is the tendency for 
standard deviations of the older age groups to grow greater, 
hence A. K's profile plotted on a form prepared from data 
secured from 14 year old children would show the profile 
line to be nearer the line of the means than it is to the broken 
line in Figure 15. While the data for 14 year old children 
are not available, the data for 13 year old children make it 
possible to plot this profile on the chart for the latter. The 
result indicates that the profile is much more even than it is 
in Figure 15. 

In the profile of Figure 15 there are numerous conspicu- 
ous differences. It appears that the score in History and 
Civics Information is much higher (about 5 S.D's) than the 
score in Computation. On the other hand, the difference be- 
tween the score in Word Meaning and that in Sentence Mean- 
ing is not large. How large must a difference be before we 
can pronounce it a true difference, independent of such fac- 
tors as the unreliability of the tests? We must answer this 
by use of the standard deviation of such differences. Kel- 
ley's formula o d . oow =V2 -rn - r 2 n supplies the standard error 
needed. These standard errors were calculated for all pos- 
sible differences between test scores for the age groups 6 to 
13 inclusive. Space does not permit the publication of these 
here. They are on file in the Library at Stanford University. 

The construction of the profiles also made necessary the 
calculation of means, standard deviations, and reliability co- 
efficients for each test for each age group. In the case of the 



THE SPECIALIZATION OF ABILITIES 343 

Stanford Achievement Tests these calculations were made 
from the groups on which the tests were standardized. 
Other groups were used for calculating the reliability co- 
efficients for the Information Tests and for the Stanford- 
Binet. These data are shown in Tables 132, 133, and 134. 

TABLE 132 

STANDARD DEVIATIONS OF AGE-GROUP DISTRIBUTION FOR TESTS 
AS INDICATED 

Age Groups 7 8 9 10 11 12 13 

Stanford-Binet 

Months 10.07 12.61 12.95 13.74 18.51 18.47 20.87 

Paragraph 

Meaning 12.85 20.00 22.80 24.00 25.25 21.20 23.00 

Sentence Meaning 7.42 10.59 13.80 14.80 18.40 18.00 19.60 
Word Meaning 6.28 11.10 14.60 16.80 18.36 16.60 18.80 

Arithmetic 

Computation 15.00 22.80 24.80 32.80 26.65 29.60 30.00 

Arithmetic 

Reasoning 9.40 13.90 17.60 19.32 21.96 25.60 28.87 

Language Usage 11.50* 11.50 8.05 11.35 13.30 12.24 14.20 
Spelling 21.25 26.73 28.20 35.08 39.08 39.40 43.10 

Science 

Information 6.17* 6.17 13.39 14.71 19.97 21.12 21.72 

Language and 

Literature 

Information 5.68* 5.68 9.33 9.87 15.90 18.30 20.12 

History and Civics 

Information 3.8* 3.80 5.83 10.13 15.26 20.37 20.67 

Music and Art 

Information 2.80* 2.80 3.67 4.31 6.41 8.40 9.03 

Information Total 15.78* 15.78 32.22 36.57 55.40 65.79 69.49 
Reading Total 26.55 41.29 49.32 56.4 60.10 55.60 61.10 

Arithmetic Total 24.00 36.66 35.97 51.2 50.80 55.20 58.40 

* Standard deviation for 8 year group used, as there were no data on un- 
selected 7 year olds. 

With the standard errors let us examine the differences 
found in Figure 15 (A. K's profile). Let us take the differ- 
ence between any two scores, such as Paragraph Meaning 
1.94 and Arithmetic Computation .38. Here the difference is 



344 TRAITS OF GIFTED CHILDREN 

TABLE 133 
MEAN SCORES FOR TESTS AND AGE GROUPS AS INDICATED 

Age Groups 78 9 10 11 12 13 

Stanford-Binet 

Years 7.5 8.5 9.5 10.5 11.5 12.5 13.5 

Paragraph Meaning 7.8 18.4 31.6 45.4 55.6 63.6 70-2 
Sentence Meaning 5.0 10.0 17.4 24.9 33.6 39.8 48.0 
Word Meaning 3.3 8.7 17.3 26.5 36.4 42.6 49.4 

Arithmetic 

Computation 16.0 34.0 49.2 71.6 86.0 101.6 108.0 

Arithmetic 

Reasoning 8.0 17.6 28.4 41.6 49.2 64.8 73.6 

Language Usage -1.0 2.0 7.0 13.0 21.0 23.8 28.4 

Spelling 24.5 39.6 56.8 74.6 81.4 110.3 129.8 

Science 

Information -1.0* 6.7 14.5 22.5 30.5 38.5 46.0 

Language and 

Literature 

Information 3.0* 8.4 13.8 18.5 23.5 30.5 37.0 

History and Civics 

Information -2.5* 3.7 7.1 13.0 19.5 26.5 33.5 

Music and Art 

Information 0.0* 3.9 6.6 8.4 11.0 14.1 16.4 

Information Total 0.0* 20.5 42.0 62.0 84.5 107.5 130.5 
Reading Total 17.0 37.5 66.5 96.5 125.5 146.0 167.5 

Arithmetic Total 24.0 52.0 78.0 112.0 136.0 166.0 182.0 

* Means estimated from a smoothed curve. 

1.56 standard deviation. We find the standard error of such 
a difference to be .37. Then the obtained difference of 1.56 

1.56 

standard deviations of the distributions is actually ^r or 

.o/ 

4.22 standard deviations of such a difference. Hence the 
probability of this difference being as small as is .00003. 
This method of reasoning may be applied to any of the dif- 
ferences encountered in such a profile. If it were applied ta 
all of the differences in each of the 550 profiles it would fill 
550 tables each having 105 entries. For this obviously too 
cumbersome procedure we may substitute rough approxi- 
mate mental calculations in our scrutiny of the profiles and 



THE SPECIALIZATION OF ABILITIES 345 

TABLE 134 
RELIABILITY COEFFICIENTS FOR AGE GROUPS AND TESTS AS INDICATED 

Age Groups 7 8 9 10 11 12 13 

Stanford-Binet M.A. .92 .92 .90 .93 .93 .94 .91 

Language Usage .76 .76 .76 .75 .77 .80 .83 

Spelling .92 .95 .92 .95 .96 .95 .96 

Computation .87 .90 .92 .92 .89 .85 .85 

Arithmetic Reasoning .87 .84 .87 .88 .87 .90 .91 

Word Meaning .94 .95 .94 .96 .92 .94 .96 

Sentence Meaning .91 .89 .89 .87 .89 .88 .89 

Paragraph Meaning .92 .95 .95 .94 .92 .90 .91 

Science Information .65 .87 .87 .90 .90 .91 

Language and Literature .67 .67 .69 .78 .87 .91 .91 

History and Civics .60 .52 .73 .81 .94 .91 

Music and Art .58 .48 .64 .72 .75 .88 

Reading .95 .95 .96 .96 .93 .96 .96 

Arithmetic .88 .93 .93 .92 .90 .93 .93 

Information .85 .89 .94 .89 .96 .91 

more precise calculations when we wish to study one case 
intensively or many cases as a composite profile. An ex- 
ample of the quick examination is furnished by attempts to 
discover the valid differences in Figure 15. A. K. appears to 
be more proficient in reading than he is in arithmetic. His 
score in Reading Total is about 1 1/3 standard deviation 
above his score in Arithmetic Total. The standard error of 
this difference is .35; hence the difference is 3 or 4 standard 
errors, which is enough to warrant its validity. With stand- 
ard errors in the neighborhood of .4 for differences be- 
tween the three reading tests, and differences in the profile 
of less than .3 z scores, we may conclude that A. K*s abilities 
are very even in Word Meaning, Sentence Meaning, and 
Paragraph Meaning. 

ACHIEVEMENT LEVELS AND INTELLIGENCE LEVELS 

The mean educational quotient of the main group of 643 
gifted children was found to be about 140, and the mean IQ 
about 150. This difference between the achievement level 
and the intelligence level of the gifted children is very sig- 
nificant if it be a true difference. The method described in 
the preceding section may be applied here to test this differ- 
ence. The procedure is simple. 

The difference between the standard score for an achieve- 



346 TRAITS OF GIFTED CHILDREN 

ment test and the standard score for the Stanf ord-Binet Test 
is expressed as positive when the Stanford-Binet score is 
higher, and negative when the Stanford-Binet score is lower 
than the Achievement score. These differences were calcu- 
lated for all of the gifted subjects. Each difference was then 

divided by the standard error of that difference (^^J 
and the results tabulated by age groups. The means of these 
results are given in Table 135. 

TABLE 135 
PROBABILITY OF DIFFERENCES BETWEEN THE STANFORD-BINET SCORES 

AND THE SCORES OF THE SEVERAL ACHIEVEMENT TESTS EXPRESSED 



IN 1KJRMS OF TJtiJi 


AVJ&ttAU 


rJa U. 


n j 




r\jx\ 


-OLJ-iJl 


J> J. JTJ. 










Age Groups 


7 


8 


9 


10 


11 


12 


13 


Language Usage 


-5.64 


-3 


.13 


-1 


.57 


-3 


.97 


-4.00 


-2 


.77 


-2.38 


Spelling 


-5.35 


-4 


.74 


-4 


.30 


-6.67 


-5.45 


-4 


.26 


-3.74 


Computation 


-3.73 


-4.63 


-5 


.51 


-7 


.40 


-4.30 


-4 


.22 


-3.21 


Arithmetic 
























Reasoning 


-2.28 


-2 


.44 


-3 


.96 


-4 


.69 


-2.35 


-3 


.38 


-2.90 


Word Meaning 


-1.92 


-2 


.99 


-4 


.95 


-6 


.70 


-4.17 


-3 


.82 


-3.51 


Sentence Meaning 


-4.23 


-3 


.09 


-4 


.60 


-4 


.60 


-3.75 


-3 


.60 


-3.48 


Paragraph Meaning 


-3.30 


-4 


.87 


-5 


.85 


-7 


.09 


-4.91 


-3 


.90 


-3.59 


Science Information 


-2.56 


+ 


.62 


-3 


.36 


-3 


.86 


-3.43 


-3 


.60 


-2.22 


Language and 
























Literature 


-2.82 


+ 


.26 


-1 


.97 


-2 


.08 


-2.20 


-2 


.76 


-2.10 


History and Civics 


-3.73 


+ 


.07 


-1 


.43 


-1 


.78 


-1.77 


-3 


.64 


-2.02 


Music and Art 


-3.35 


- 


.59 


-1 


.45 


-1 


.22 


-1.10 


-2 


.27 


-1.26 


Reading Total 


-3.78 


-4 


.02 


-5 


.89 


-7 


.02 


-4.51 


-4 


.62 


-3.80 


Arithmetic Total 


-3.54 


-4 


.29 


-5 


.12 


-6 


.50 


-3.56 


-4 


.56 


-3.39 


Information Total 


-3.04 


+2 


.76 


-2 


.90 


-3 


.16 


-2.43 


-3 


.62 


+ .74 



The data of Table 135 agree in general with the results 
of the use of the educational quotients in their indica- 
tion that the intelligence level of the gifted children is higher 
than their achievement level. There are 77 differences in the 
table if we omit the differences involving total scores. Of 
these, 73 are negative and greater than one standard devia- 
tion of such a difference. We may make a more exact com- 
parison. 



THE SPECIALIZATION OF ABILITIES 347 

Since the educational quotient was found to be about 140, 
and the IQ about 150, it may be stated that the achievement 
level of the gifted children is 40 per cent above that ex- 
pected for their chronological ages and approximately 10 
per cent below that expected for their intelligence level. We 
find the mean of all the Stanford-Binet standard scores is 
3.98, and the mean of all the achievement test standard 
scores is 2.47. The score 2.47 is 62 per cent of the 3.98, 
which is the difference between the intelligence level of un- 
selected children and the intelligence level of gifted chil- 
dren, as expressed in terms of standard scores. The differ- 
ence between the mean EQ and the obtained EQ is 40, which 
is 80 per cent of the difference between the intelligence levels 
of unselected children and of gifted children expressed in 
terms of quotients. On the basis of EQ the gifted children are 
achieving 80 per cent of the excess achievement which their 
intelligence would warrant; on the basis of standard scores 
62 per cent. While the gifted children have an achievement 
level well above that of other children, it is nevertheless so 
much below their level of intelligence as to make the condi- 
tion a definite challenge to both the parent and the teacher. 

By use of the standard error we can give additional con- 
firmation to the results just described. Again weighting all 
the tests equally we find their mean reliability coefficient to 
be .829. This reliability coefficient makes possible the calcu- 
lation of the standard error of the difference (a&oo), which is 
.5848 z. Hence the probable verity of the difference 1.51 z is 

1.51 

-7^- or 3.18 CA.OOW Then the chances of the difference be- 
* .5848 

tween the achievement level and the intelligence level being 
as small as are not over .0007. 

It might be asserted that these gifted children represent 
a highly selected group whose Stanford-Binet scores are 
spuriously high. That is, by the method of selection those 
who accidentally made low Stanford-Binet scores were re- 
jected and those whose intelligence is actually lower than 
the Stanford-Binet scores indicate, secured high scores be- 
cause of the less than perfect reliability of the Stanford- 
Binet test. The correction 1 which would apply to these scores 

a Kelley, T. L. Statistical Methods, The Macmillan Co. (1923), Sec. 60, 
p. 214 ff. 



348 TRAITS OF GIFTED CHILDREN 



would then be: X^=r u ^ + (1 r n ) M, in which X^ is the 
true score, rn the reliability coefficient, X x the obtained 
score, and M the mean of the obtained distribution. But 
when the reliability coefficient is .90 or more the correction 
is negligible. Hence if we assume that each gifted child is a 
part of a distribution of the age group, these corrections 
would all be too small to change our conclusions, as the re- 
liability coefficients for the Stanford-Binet for age groups 
are all in the neighborhood of 90 or better. But it might be 
assumed that the gifted children are to be treated as sepa- 
rate distributions selected by means of the Stanford-Binet 
test. In this case the correction would be in the direction of 
the means of the age groups of the gifted children, and the 
reliability coefficients for such groups would be used. The 
mean mental ages of the age groups of gifted children are : 

Age group M.A. 

7 11.10 

8 13.02 

9 14.36 

10 15.57 

11 16.92 

12 17.62 

13 18.58 

We have no reliability coefficients for M.A. distributions of 
children. It is very possible that a group whose M.A. scores 
were between 13-0 and 13-11 would yield a much lower re- 
liability coefficient than the r=.905 which was found for the 
chronological age distribution 13-0 to 13-11 with a normal 
distribution of IQ's. If the reliability of M.A. groups were 
the same as the reliability of chronological age groups, then 
we know the corrections would be of no significance. 

But zero is the theoretical limit of a reliability coefficient. 
Assuming that the reliability coefficients of the M.A. groups 
were all as low as zero, the mental age scores would be, when 
corrected, as low as the means of their respective mental 
age groups. It would then be logical to compare the mean 
standard scores for the Stanford-Binet of each age group 
with the mean standard scores of the other tests. But it 
was by this comparison that we concluded that the intelli- 



THE SPECIALIZATION OF ABILITIES 349 

gence level of the gifted children is 1.51 standard scores 
above the achievement level. 

Two facts stand out in this study: (1) that any adequate 
conception of the surface of the abilities of these gifted chil- 
dren can be gained only by the most careful scrutiny of the 
profile charts of individual children; (2) that the gifted chil- 
dren have an intelligence level which is considerably above 
their achievement level. 

CASE STUDIES 

The following case studies are not complete pictures of 
the characteristics of the children studied. They do not uti- 
lize all of the data available. They are presented merely to 
indicate the possibilities of this method and to show certain 
interesting and valuable facts. The nine cases presented 
here are totally inadequate to represent the great variety of 
profiles which were found in the whole survey. More than 
five hundred of these profiles were drawn and studied. These 
are on file at the Library and with the Department of Psy- 
chology at Stanford University. 

The cases given here represent a number of age groups 
and in part are examples of rather extreme specialization in 
some one ability. Wherever possible, data from parents and 
teachers were examined to corroborate or qualify the find- 
ings of the tests and the treatment of the scores. 

H. M. J. (Figure 16), who was 6 years, 9 months old when 
she scored M.A. 13 on the Stanford-Binet, thus making an IQ 
of 192, had the highest IQ in the 6 year group. This was just 
a little lower than the highest IQ found in the group of 643 
children. Because of the small number of 6 year old chil- 
dren who can make scores on the Stanford Achievement 
tests, there are no valid means and standard deviations 
available, hence the 7 year profile chart is used. Although 
not 7 years old, H. M. J. is in the high third grade. Her 
Arithmetic Reasoning standard score is the highest of her 
Achievement scores, and that in Music and Art Information 
standard score the lowest. The difference between her Arith- 
metic Reasoning score and her Reading Total score is 2.06 
z scores. The standard deviation of this difference (ad.ooto 
from Table 132) is .46. Then this difference is 4.48 standard 



350 



TRAITS OF GIFTED CHILDREN 



deviations. H. M. J's abilities in Arithmetic Reasoning, in 
Word Meaning, and in Paragraph Meaning are unusual. Her 
Spelling and Computation are on about the same level in the 
middle range of her abilities. History and Civics, and Music 
and Art Information are low. For so young a child, with so 

FIGURE 16 
PROFILE OF H. M. J. 

(Girl; age, 6-9; mental age, 13-0; IQ, 192; grade, high 3.) 



<o ti c 6 c: c C .- .0 o /> ~" ~Z - 

! > > ^^SSsSo!s < rc*| H 


< c a s s v ~ ,0 ,2 , 


is>i-??~ivi;^-< 


23<ll<3c8fl^c*<.! 






























































(A c 








/ 


" - ' - 


\ 














s 




?) G 


\ 




> 






\ 














f 


\ 


c 


V 




y 




















f S 


\ 


o 

05 
CO 


\ 




/ 


/ 


\ 




X 










5 




\ 

\. 


\ 

\ 
1 


1 




/ 


\ 






\ 

X 












-o 

(TJ _ 


\ 

\ 


/ 




/ 




\ 


/ 


\\ 








4 


^ 


\\ 


o 3 

<0 

*- 9 






X 






\ 


/ 


\ 


" v -^ 


*e^ 








^ 


co c 


















** ^ Jr " 


\ 




2 




























V 


1 








form 


/Sr>< 


vc 


^ 












\ 




























\ 









Solid llae, obtained standard scores. 
Broken line, age equivalents for mental age. 

brief a school experience, this amount of unevenness is sig- 
nificant and suggests real innate differences. 

The evidence from the profile which points toward spe- 
cialized abilities in Arithmetic and Reading are supported by 
the parents* account of the early childhood of H. M. J. She 
learned to read when 3 years old. Before she was 6 years 



THE SPECIALIZATION OF ABILITIES 



351 



old her father states' that "she carried the powers of 2 men- 
tally up to the 20th power (1,048,576) as a Sunday afternoon 
pastime." He thinks she could have carried this farther but 
he feared the practice would tire her. She could count to 100 
when 3 years old. 

FIGURE 17 
PROFILE OF N. M. 

(Boy; age, 7-11; mental age, 13-8; IQ, 173; grade, high 3.) 




Solid line, obtained standard scores. 
Broken line, age equivalents for mental age. 

N. M. (Figure 17) was 7 years, 11 months old when he 
scored an IQ of 173, which is the highest score in the 7 year 
old group. Like the preceding case, N. M. is a "reasoner." 
His score in Arithmetic Reasoning is phenomenally high. His 
teacher also reports that he is superior in Arithmetic. She 
makes the same estimate of his reading ability, but his Arith- 



352 TRAITS OF GIFTED CHILDREN 

metic Total score is 3.11 standard scores, or 7.58 S.D's above 
his Reading Total score. This difference is substantial. 

N. M's teacher ranks him only a little below average in 
Composition. His profile indicates that his Language Usage 
score is very low as compared with the other scores. It is 
1.89 standard errors below his Reading Total score, and 8.44 
standard errors below his Arithmetic Total scores. Other low 
points in his profile are Spelling, Sentence Meaning, and 
Music and Art Information. The evenness of the Paragraph 
Meaning, Science Information, and Language and Literature 
Information is conspicuous. Spelling and Computation dif- 
fer 1 z score or a little over 2 standard errors. 

This boy, not yet 8 years old and in the third grade, 
shows a degree of unevenness greater than we can account 
for by the training he has received in his brief school experi- 
ence. The parents report no home tutoring other than the 
usual incidental instruction. The differences in ability in 
this boy's case seem to be to a large extent due to native en- 
dowment. 

S. M, F. (Figure 18) was 9 years and 6 months old when 
she made an IQ of 189, the highest for her age group. Her 
profile is characterized by more than usual evenness among 
the achievement scores and a very unusual difference be- 
tween the achievement and M.A. levels. The greatest differ- 
ence in her profile, that between Language and Literature 
Information and Paragraph Meaning, is so situated as to be 
rather noticeable. This difference is 2.33 standard scores or 
3.88 standard deviations of this difference. On the other 
hand, the Language and Literature Information and the 
History and Civics Information are close together, the dif- 
ference being but .36 standard scores. The S.D. of this 
difference is but .4. That is, S. M. F. shows the greatest 
abilities in these two information tests. With an IQ of 189, 
her Reading quotient is 163 and her Arithmetic quotient is 
151. In terms of standard scores her achievement scores 
are all more than two standard deviations above her age 
expectancy. Hence, although her mental level is far above 
her achievement level, her achievement level is conspicu- 
ously above that of children of her age. 

The zigzag contour of this profile suggests that the abilities 
measured have not developed evenly. The following differ- 
ences with their standard errors confirm this assumption: 



Spelling 
Language Usage 

Sentence Meaning 
Language and 
Literature Information 

Language and 
Literature Information 

Language and 
Literature Information 

Sentence 
Meaning 



c 

0) 



- 



3.11 
3.74 

4.63 
4.63 
4.63 

3.74 

FIGURE 18 
PROFILE OF S. M. F. 

nental age, 17 
c c 






OF ABILITIES 


353 


Comp. 


2.53 


= 1.20 


fffcoOID - 4 <> 


<t 


2.53 


= .58 


" .57 


tf 


2.53 


= 1.21 


" .44 


<( 


2.53 


= 2.10 


" .62 


Lang. 
Usage 


3.11 


= 1.52 


" .74 


Par. 








Mean. 


2.30 


= 2.33 


" .60 


Par. 








Mean. 


2.30 


= 1.44 


" .40 


M. F. 








IQ, 189; grade, high 5.) 


k 


c 


L 


^ _ 


V. *c 


c2 


^ flj 


4? "5 





*1 


*-* "o 


1 1 


i-* Hi 

fr 

CO JJ 


o 
* 


1 I 

1 Of 


1 1 

< - 





2. 


j 


J > C 


J> < 


i : 


& I 


( 


L < 


. 


-1 \ 


< 


C 


^ 


t " 






















1 

1 


\ 










I 




















\ 


8 








A\ 


















1 


\ 






/ 







\ 




/ 


\ 








y 


I 




6 




/ 


O 5 
co 


\ 




"^*. 


/ 


\ 





s 


/ 





' 




5 


/ 


y 


4 

5-r 


^ 


\^ 


\ 


/ 


X 


X 


V s 


/ 


^ 




\ 


4- 




/ 


u 3 

<0 

4- n 






\ 


/ 






\ 


/ 








3 


^ J 


































1 


/4^e/ 


Torrn 


fy-e 


year 
























.4 































Solid line, obtained standard scores. 
Broken line, age equivalents for mental 



354 



TRAITS OF GIFTED CHILDREN 



These are but a small sampling of the measurable dif- 
ferences shown in Figure 18. 

S. D's quotient of 175 (Figure 19) is the highest IQ of the 
10 year old group of gifted children. Only in the case of the 
score in Computation does her achievement score fall below 

FIGURE 19 

PROFILE OF S. D. 

(Girl; age, 10-3; mental age, 17-11; IQ, 175; grade, high 5.) 



42 

Q> 






fi- 



o 

I 



m 

oe 







r 






i 

<D 



I I 



I 



f 1 * * 

W J 



I 



IQ 

f I 

i J> 



! 


O5 

CO 



rt. 
7J< 

l< 



^^g / toW7 6r /<^ ygoy ? 









1 


-1 

Solid line, obtained standard scores. 
Broken line, age equivalents for mental age. 

the level of 2 standard scores above the mean for 10 year old 
children. As might be expected from her high IQ, her intelli- 
gence level is markedly above her achievement level. 

Her scores are quite even in the Reading tests. The Arith- 
metic scores, and the Spelling score are not very uneven nor 
do they differ very much from the Reading scores. The 



THE SPECIALIZATION OF ABILITIES 355 

largest difference in this group of scores is that between 
Sentence Meaning and Computation, which is 1.24 standard 
scores, or 2.7 times the standard deviation of this difference. 
This is a real difference. The scores in Language and Lit- 
erature (4.41), in History and Civics (3.45), and in Science 
Information (3.50), stand out decisively above the level of 
the other achievement group (Arithmetic, Reading, and 
Spelling) . 

The parents and teacher of this child report her to be 
very superior in Music. The Music and Art Information Test 
does not reveal a superiority over the other abilities meas- 
ured, although it does rate her 2.69 standard deviations 
above the mean for children of her age. 

In addition to earning the highest IQ made by the 11 year 
old gifted children, A. A. (Figure 20) made the highest pos- 
sible score in the Spelling Dictation test. Her score in History 
and Civics Information is 86 out of a possible 90; in Music 
and Art Information it is 42 out of a possible 45; in Lan- 
guage Usage it is 54 out of a possible 60. In these tests it is 
probable that her ability has not been fully measured. 

A. A's profile shows a real superiority in the ability meas- 
ured by the Music and Art Information Test. This is shown 
by the differences found when the score on this test is com- 
pared with the Reading Total and Arithmetic Total. These 
differences are as follows: Music and Art score is 1.74 z 
scores or 2.95 S.D. higher than the Reading Total, and 2.27 
z scores or 3T66 S.D. higher than Arithmetic Total. Both her 
mother and her teacher report A.A. as superior in musical 
ability and the girl herself reports that she likes music very 
much. She plays the 'cello. 

A. A's teacher also reports her mental abilities as "very 
even," although we find substantial unevenness in her pro- 
file. For example, in addition to those just given, there is a 
difference between Arithmetic Reasoning and Paragraph 
Meaning of 1.98 z scores or 4.3 S.D. The probability of this 
difference not existing may be expressed as .00002. The dif- 
ferences between Arithmetic Reasoning and Language 
Usage, and between Word Meaning and Sentence Meaning, 
are also substantial. If the teacher were comparing A.A. 
with other children, the unevenness of her abilities might 
not be prominent, but in the abilities measured by these tests 



356 



TRAITS OF GIFTED CHILDREN 



her abilities are far from even. Evidence of unevenness not 
measured is found in her statements that she dislikes very 
much freehand drawing, painting, folk dancing, manual 
training, and mechanical drawing. Her teacher rates her as 
below average in penmanship and manual training. 

FIGURE 20 

PROFILE OF A, A. 

(Girl; age, 11-11; mental age, 20-4; IQ, 171; grade, low 9.) 

w <0 ecocd,-po,o "T 

4> & w 0'0 < o<a c ^'~ < *** 

*F^ : i.3!z!*f:j3i5! 

1 * ? I "S f^^t"^^i 

ft)^Q.i'rp <D 1; CT 13 ? Q> *- e 

2.3<o<3<:i<ocL(o_5x:EC*<>5 



y 




























































7 

ift c 






























S G . 

f s ( 


L 




























CO 


Y 
\\ 






X 


\ 
\ 
V 










/ 


s 






/ 


C 


\\ 


i"- 


^** 


/ 


V 


,--- 


- N 




. - 


's' 




4 


S 


> 


(Q 


\ 


/ 






\ 




V 


r~^f 










^^^ 


/ 


C 


^ 


/ 






\ 






/ 








3 


^^x 




5 














"^s^ 


/ 














c/; <- 
























Z 








^ 


orm 


/&// 




\s 














1 





































Solid line, obtained standard scores. 
Broken line, age equivalents for mental age. 

A.A. says she rather dislikes Spelling but finds it the easi- 
est subject. As she made the highest possible score in this 
test it is probable that she has an ability in Spelling not 
shown by the profile chart. 

G. F. (Figure 21), has the highest IQ found in the 12 year 
old group. Her profile is more even than those for the 



THE SPECIALIZATION OF ABILITIES 



357 



younger age groups, partly because the scores are nearing 
the upper limits of the tests, and partly because the standard 
deviations are larger for these older children. On the other 
hand, the standard errors of the type a&oow are smaller for 
the 13 year olds than for the younger age groups. 

FIGURE 21 
PROFILE OF G. F. 

(Girl; age, 12-1; mental age, 18-11; IQ, 157; grade, low 8.) 



. 



^ , 







rt 

D 






Ae 



\/ 






Solid line, obtained standard scores. 
Broken line, age equivalents for mental age. 

The difference between her Arithmetic Reasoning and 
Science Information scores is 1.54 z scores and 3.42 S.D. The 
difference between Language and Literature Information 
and Science Information is 1.3 z scores or 2.95 S. D. These 
differences are substantial and indicate a specialization in 
Arithmetic Reasoning and in Language and Literature In- 



358 



TRAITS OF GIFTED CHILDREN 



formation. Accompanying this superior ability in Language 
and Literature Information there is very superior ability in 
Reading, for the Paragraph Meaning score is the maximum 
and the Reading Total score is very near the maximum for 
the tests. Her teacher rates her as superior or very superior 

FIGURE 22 
PROFILE OF Z. J. 

(Boy; age, 10-7; mental age, 17-0; IQ, 161; grade, low 7.) 




Solid line, obtained standard scores. 
Broken line, age equivalents for mental age. 

in all of the school subjects. With this rating, with an 
achievement level so high, and with an M. A. score of 18.11, 
one cannot but wonder why G. F. should be in the low 
eighth grade. 

Z. J's profile (Figure 22) shows unusual specialization in 
the Information scores. His score in Music and Art Informa- 



THE SPECIALIZATION OF ABILITIES 



359 



tion stands out above all other scores. His father plays, 
sings, and draws. This home influence is doubtless reflected 
in the high score. Z. X finds Drawing the easiest of all his 
studies. He thinks he will be an architect but may decide to 
be an advertiser, decorator, cartoonist, magazine illustrator, 

FIGURE 23 
PROFILE OF D. E. R. 

(Girl; age, 8-9; mental age, 13-7; IQ, 155; grade, low 4.) 






Solid line, obtained standard scores. 
Broken line, age equivalents for mental age. 

artist (painter of pictures), landscape artist, civil engineer, 
chemist, zoologist, etc. All of which supports the evidence of 
this profile chart. 

This little girl (Figure 23) shows prominent specializa- 
tion in Science Information, and in Language and Literature 
Information. Her ability in Reading is well above the mean 
ability for her age and grade. Her Reading scoresr are sub- 



360 



TRAITS OF GIFTED CHILDREN 



stantially higher than her scores in Arithmetic Computation, 
Arithmetic Reasoning, and Spelling. There is a marked 
superiority of her abilities in Language Usage and Reading 
when they are compared with her abilities in Arithmetic. 
The reports of her teacher and of her parents both indicate 

FIGURE 24 
PROFILE OF J. D. 

(Boy; age, 10-3; mental age, 15-11; IQ, 155; grade, high 5.) 









I 



(0 

-a 

c 




Solid line, obtained standard scores. 
Broken line, age equivalents for mental age. 

an ability in Arithmetic much lower than the abilities in 
other school subjects. 

Her superior score in Science Information is doubtless 
due to her early and persistent interest in birds, animals, 
flowers, and the out-of-doors. Her interest in Literature and 
her Reading scores are due in part to her early training. 



THE SPECIALIZATION OF ABILITIES 361 

She learned to read before going to school. Her parents re- 
port some early instruction in Reading, and rather more 
than the usual amount of time spent in reading to her. 

J. D. This boy, like D. E. R., has a low level of perform- 
ance in Arithmetic. The Arithmetic Total score is 1.94 stand- 
ard deviations below his Reading Total score. The standard 
error for this difference is .35, hence the difference is equal 
to 5.54 S.D. The difference between the Arithmetic Total 
arid the Information Total is even greater. If all of the 105 
differences were calculated for this profile they would fur- 
nish convincing evidence of the unevenness of the abilities 
as measured. 

This profile of J. D. reveals two levels of scores. Spelling, 
Computation, and Arithmetic Reasoning are on one level, 
which is the lower level ranging from .75 z scores to 1.5 z 
scores. The three reading scores and the Language Usage 
score are between two and three standard scores above the 
means, making a second level. Then ranging above these 
levels is a series of steps ascending from Science Informa- 
tion, 3.02, to Language and Literature Information, 3.49, to 
Stanf ord-Binet, 4.73, to History and Civics Information, 5.53, 
and to Music and Art Information, 6.4. 

SUMMARY AND CONCLUSIONS 

The following statements are the outcome of the investi- 
gations described in the preceding pages: 

1. The accumulation of improved instruments for the 
measurement of human traits, and the refinement of statis- 
tical procedure, make possible a new attack on the problem 
of the differences between traits resident in an individual. 

2. Both the gifted children and the unselected children 
who were investigated, show such real and varied differ- 
ences between their abilities in school subjects as to warrant 
the statement that each child must be regarded as a unique 
individual with specific mental mechanisms. 

3. When a gifted group is compared with various sample 
groups of normal children, no large differences in the in- 
equalities of their levels of achievement are discoverable. 
The gifted children have unevenness of the surfaces of their 
abilities on a very much higher level than other children. 



362 TRAITS OF GIFTED CHILDREN 

The differences in unevenness from child to child are far 
more significant than the differences between the central 
tendencies of unevenness of any two groups. 1 

4. The best conception of the unevenness of the abilities 
of a child is secured by the careful examination of the array 
of standard scores made by that child. This examination is 
made much more significant if the observer uses the stand- 
ard deviations of the differences between standard scores 

(Cfooow?)' 

5. The gifted children selected for case studies show 
several examples of definite specialization. They strongly 
suggest that the extension of this method by the use of other 
tests of other abilities will furnish invaluable assistance in 
vocational and educational guidance. 

6. There is a significant lack of parallelism in the devel- 
opment of abilities to deal with school subjects. This 
unevenness of development in the cases of the younger chil- 
dren indicates differences which are too great to be ac- 
counted for by the differences in training. The development 
of some of the mental abilities must be greatly facilitated by 
innate factors. 

7. The scores in Arithmetic Reasoning, Arithmetic Com- 
putation, and Spelling Dictation are conspicuous in their 
ability to preserve their independence from other test scores. 

8. It is definitely established that the intelligence level of 
the gifted children is higher than the level of their achieve- 
ment scores. The use of the quotients, the use of standard 
scores, and the use of the standard errors (a*. cow) all agree 
on this point. The quotient method and the method of stand- 
ard scores are in fair agreement in gauging the relation be- 
tween the intelligence and the achievement levels. If the 
altitude of the intelligence level above the age norm is the 
base, the altitude of the achievement level above its age 
norm is found to be 62 per cent by use of the standard 
scores and 80 per cent by use of the educational quotients. 

9. The abilities in the school subjects measured are more 
responsive to the influence of a general ability as measured 
by the Stanford-Binet Scale than they are responsive to any 
specialized ability measured at present. 

i See pp. 549-550 for explanation of the slight difference that may exist 
between gifted and unselected children in unevenness of abilities. L. M. T. 



CHAPTER XIII 

SCHOLASTIC, OCCUPATIONAL, AND 
OTHER INTERESTS 1 

A four-page Interest Blank, 8% x 11 inches, was filled out 
by the children of all the gifted groups and by control group 
A. 2 The first page called for ratings on the different school 
subjects according to preference and ease; the second page 
contained a list of 125 occupations to be checked so as to in- 
dicate vocational preference; the third and fourth pages 
were devoted to data on reading interests, collections, inter- 
est in various kinds of activities, and records of accomplish- 
ments and of offices and honors held. This chapter will 
summarize for the main gifted group and for the control 
group the data on scholastic interests, occupational inter- 
ests, collections, and interests in activities. 

SCHOLASTIC INTERESTS 

Ratings of the various school subjects according to pref- 
erence were obtained from the gifted and control groups ac- 
cording to the instructions printed here in reduced type: 

FIRST, DRAW A LINE RIGHT THROUGH EACH SUBJECT YOU HAVE NEVER STUDIED. 

1. Next, put a figure 1 on the dotted line before each subject that you LIKE 
VERY MUCH. 

2. Put a 2 before each subject that you LIKE FAIRLY WELL. 

3. Put a 3 before each subject that you NEITHER LIKE NOR DISLIKE. 

4. Put a 4 before each subject that you RATHER DISLIKE. 

5. Put a 5 before each subject that you DISLIKE VERY MUCH. 

Now, put ONE CROSS, like this, X, before each subject that is VERY EASY FOR 
YOU; put TWO CROSSES, XX, before the ONE subject that is EASIEST 
OF ALL. 

ART, ETC. FOREIGN LANGUAGE PRACTICAL SUBJECTS 

Drawing (free hand) French Agriculture 

Modeling German Bookkeeping 

Music (instrumental) Latin Cooking 

Music (singing) Spanish Manual training 

Painting, water Mechanical drawing 

colors, etc. HISTORY Sewing 

Ancient or medieval Shop work.tools, etc. 

Civics or citizenship Typewriting 

United States history 

1 Written with the assistance of Jennie Benson Wyman. 
2 See pp. 177-178 for description of this control group. 

363 



364 TRAITS OF GIFTED CHILDREN 

ENGLISH MATHEMATICS SCIENCE 

Composition Algebra Botany 

Debating or speaking Arithmetic Chemistry 

...Dramatics Geometry General science 

...Grammar Trigonometry Geography 

....Literature .Nature study 

Penmanship PHYSICAL EDUCATION Physical geography 

.... JReading Folk dancing Physics 

Spelling Games and sports Physiology or hygiene 

Military training Zoology 

Physical training 

As many of the subjects had not been studied by all the 
children, the number at a given age who rated a subject was 
often small. Ages 11, 12, and 13 have therefore been com- 
bined. Ages below 11 are omitted from the following tables 
because of the frequency with which the younger children of 
the control group were unable to understand and carry out 
the instructions. Table 136 gives the means and the dis- 
tributions of ratings by subject for the gifted and control of 
ages 11, 12, and 13. 

The numbers were as follows : 

Boys Girls 

Gifted Control Gifted Control 



61 


50 


44 


40 


38 


38 


40 


28 


11 


49 


15 


59 



Total 110 135 99 127 



TABLE 136 
SCHOOL SUBJECT INTERESTS OF GIFTED AND CONTROL GROUPS 

(Ages 11, 12, and 13 combined) 

Ratings by Boys Ratings by Girls 

Subject 1234 5 M 123 45M 

Drawing 

Gifted 32 38 16 10 5 2.18 39 28 11 9 1 1.92 

Control 44 42 23 4 6 2.04 46 34 17 4 3 1.88 

Modeling 

Gifted 17 13 8 3 1 2.00 24 20 6 1 1.71 

Control 54343 2.79 55222 2.43 

Music (Instrumental) 

Gifted 27 18 7 3 1 1.80 41 23 8 1 1 1 62 

Control 42 13 8 2 6 1.83 43 14 3 3 1 1.52 



SCHOLASTIC AND OTHER INTERESTS 



365 



TABLE 136 Continued 



Ratings by Boys 


Ratings by Girls 


Subject 


1 2 


3 


4 


5 


M 


1 


2 


3 


4 


5 


M 


Music (Singing) 
























Gifted 


18 35 


28 


7 


8 


2.50 


31 


33 


16 


3 


1 


1.93 


Control 


35 38 


18 


7 


14 


2.35 


66 


28 


12 


4 





1.58 


Painting 
























Gifted 


23 40 


16 


9 


7 


2.34 


41 


27 


13 


7 


3 


1.95 


Control 


35 33 


14 


4 


2 


1.92 


27 


26 


14 


2 


1 


1.91 


Composition 
























Gifted 


31 25 


29 


7 


4 


2.25 


44 


33 


13 


4 





1.75 


Control 


31 37 


13 


6 


7 


2.16 


34 


41 


23 


9 


10 


2.32 


Debating 
























Gifted 


29 17 


5 


1 


5 


1.88 


25 


16 


3 


1 


2 


1.70 


Control 


10 10 


9 


6 


6 


2.71 


8 


7 


12 


4 


6 


2.81 


Dramatics 
























Gifted 


22 10 


7 


4 





1.84 


39 


7 


1 








1.19 


Control 


5 


5 


2 


5 


3.12 


10 





1 


5 





2.06 


Grammar 
























Gifted 


26 38 


20 


12 


6 


2.35 


35 


34 


12 


8 


3 


2.02 


Control 


26 37 


23 


10 


7 


2.37 


33 


29 


20 


4 


5 


2.11 


Literature 
























Gifted 


50 33 


7 


2 


1 


1.61 


55 


19 


7 








1.41 


Control 


27 18 


16 


3 


3 


2.06 


23 


24 


6 


4 


2 


1.95 


Penmanship 
























Gifted 


7 25 


31 


26 


19 


3.23 


25 


23 


27 


10 


9 


2.52 


Control 


38 33 


23 


8 


11 


2.30 


45 


35 


16 


7 


8 


2.08 


Reading 
























Gifted 


81 20 


5 








1.28 


81 


9 


6 








1.22 


Control 


65 36 


13 


2 


3 


1.67 


68 


29 


10 


6 


4 


1.71 


Spelling 
























Gifted 


47 37 


17 


3 


1 


1.80 


45 


29 


14 


4 


3 


1.85 


Control 


50 37 


11 


5 


6 


1.90 


71 


34 


7 


2 


3 


1.56 


History (Ancient or Medieval) 


Gifted 


53 29 


5 


6 


2 


1.68 


51 


22 


7 


3 


1 


1.58 


Control 


25 17 


5 


4 


1 


1.83 


14 


6 


15 


3 


2 


2.33 


Civics or Citizenship 
























Gifted 


22 21 


20 


3 


1 


2.10 


15 


22 


18 


1 


1 


2.14 


Control 


19 24 


9 


4 


4 


2.17 


10 


20 


16 


6 


4 


2.54 


U. S. History 
























Gifted 


62 44 


7 


1 


1 


1.56 


37 


38 


13 


6 





1.87 


Control 


57 32 


10 


3 


2 


1.66 


35 


26 


17 


6 


12 


2.31 


Arithmetic 
























Gifted 


51 32 


12 


5 


7 


1.93 


45 


24 


10 


10 


5 


2.00 


Control 


53 34 


20 


11 


7 


2.18 


57 


32 


13 


7 


5 


1.87 


Folk Dancing 
























Gifted 


2 5 


7 


4 


8 


3.42 


30 


13 


11 


2 


1 


1.79 


Control 


6 3 


3 


2 


1 


2.27 


24 


8 


5 


2 


1 


1.70 



366 



TRAITS OF GIFTED CHILDREN 



TABLE 136 Concluded 



Ratings by Boys 


Subject 


1 


2 


3 


4 


5 


M 


Games and Sports 














Gifted 


80 


14 


3 








1.21 


Control 


84 


11 


3 


2 





1.23 


Physical Training 














Gifted 


45 


40 


13 


3 


1 


1.77 


Control 


52 


26 


5 


2 


9 


1.83 


Agriculture 














Gifted 


21 


19 


13 


5 





2.03 


Control 


9 


7 


6 


1 


1 


2.08 


Cooking 














Gifted 








. 











Control 




















Manual Training 














Gifted 


51 


31 


9 


4 


2 


1.71 


Control 


58 


17 


2 


1 





1.31 


Sewing 














Gifted 


- 

















Control 




















Shop Work 














Gifted 


37 


15 


4 


3 


2 


1.66 


Control 


25 


6 


6 


1 





1.55 


General Science 














Gifted 


20 


16 


6 


1 





1.72 


Control 


23 


8 


2 


1 


2 


1.36 


Geography 














Gifted 


41 


41 


19 


4 


2 


1.93 


Control 


60 


29 


10 


3 


5 


1.73 


Nature Study 














Gifted 


32 


27 


12 


1 


1 


1.79 


Control 


28 


13 


7 


1 


2 


1.75 


Physiology and Hygiene 


Gifted 


10 


23 


20 


5 


4 


2.52 


Control 


17 


17 


7 


3 


6 


2.28 



Ratings by Girls 
123 4 5 M 

63 20 5 1 1.37 

76 8 3 18 1 1.68 



36 37 17 
45 23 8 



3 1 1.89 
2 2 1.66 



10 4 4 1.67 

54211 2.15 

54 24 4 4 1 1.55 

33 19 12 1 1 1.76 



34 32 10 10 3 2.06 
39 26 8 7 3 1.90 



16 14 12 1 1 2.02 

18 12 9 1 1.85 

38 31 16 7 1.91 

41 36 16 9 4 2.05 

31 23 3 3 1 1.69 

19 12 6 3 1.83 

14 20 17 8 3 2.45 

17 13 7 5 4 2.26 



Table 137 gives the ranks of the school subjects for 
gifted and control groups. For each sex the order shows the 
relative (not the absolute) degree of preference accord- 
ing to intelligence, the subjects at the top being relatively 
most liked by the gifted, those at the bottom by the control. 
An examination of this table will show that the subjects 
relatively more liked by the gifted are those demanding the 
largest amount of abstract thinking, and that those relatively 
more liked by the control are those demanding the least. 
However, certain subjects are found in an unexpected posi- 



SCHOLASTIC AND OTHER INTERESTS 



367 



tion. For example, arithmetic, general science, and spelling 
are relatively more liked by control than gifted girls. To a 
less extent this also holds true for grammar. Cooking and 
games and sports are slightly more liked by gifted girls. 
Turning to the boys, geography, physiology and hygiene, 
and general science are relatively more liked by the control, 
and composition slightly more. Reading and arithmetic are 
only slightly better liked by the gifted, while grammar, 
games and sports, United States history, and shop work do 
not differentiate between the groups. 



TABLE 137 
RANKS OF SCHOOL SUBJECTS ACCORDING TO PREFERENCE 



Subject 



Ranks Hanks Differ- 
by by ence 
Gifted Control C-G 



Relatively more liked by G. boys 

Dramatics 13 27 14 

Debating 14 25 11 

Literature 4 * 15 11 

Modeling 17 26 9 

Little difference 

Reading 264 

Ancient History 6 10 4 

Arithmetic 15 19 4 

U.S. History 35 2 

Phys. Training 9 11 2 

Games & Sports 110 
Spelling 12 12 

Grammar 23 24 -1 

Civics 19 18 -1 

Shop Work 54-1 

Singing 25 23 -2 

Agriculture 18 16 -2 
Nature Study 10 8 -2 

Instru. Music 11 9 -2 

Folk Dancing 24 20 -4 
Composition 21 17 -4 

Relatively more liked by C. boys 

General Science 83-5 
Physiol. & Hyg. 26 21 -5 
Penmanship 27 22 -5 
Man. Training 72-5 
Drawing 20 14 -6 

Geography 16 7 -9 

Painting 22 13 -9 



Subject 



Ranks Ranks Differ- 
by by ence 
Gifted Control C-G 



Relatively more liked by G. girls 

Ancient History 6 24 18 

Debating 10 27 17 

Dramatics 1 17 16 

Modeling 11 25 14 

Agriculture 8 20 12 

Literature 4 15 11 

Composition 12 23 11 

U. S.History 15 22 7 

Reading 275 

Little difference 

Cooking 583 
Games & Sports 352 

Civics 25 26 1 

Nature Study 990 

Geography 17 6 -11 

Grammar 22 19 -3 



Relatively more liked by C. girls 

Physiol. & Hyg. 26 21 -5 

Painting 20 14 -6 

Drawing 18 12 -6 
Instru. Music 71-6 

Folk Dancing 13 6 -7 

Penmanship 27 18 -9 

Arithmetic 21 11 -10 

Sewing 24 13 -11 

Spelling 14 2 -12 

Phys. Training 16 4 -12 

General Science 23 10 -13 

Singing 19 3 -16 



368 TRAITS OF GIFTED CHILDREN 

Table 137 yields the following rank-order correlations : 
Gifted boys vs. control boys, rho -.717 



girls " " girls, 
" boys " 
" girls " " boys, 

" boys " gifted girls, 

Control " " control " 



-.165 
=.232 
-.169 
-593 
=.416 



That is, gifted boys and control boys show fairly close 
agreement (.717), while gifted girls show very little agree- 
ment with either sex of the control group. On the other 
hand, gifted girls show considerable agreement with gifted 
boys (.593). 

On page 261 are given the rank orders of the school sub- 
jects based on teachers' ratings on quality of school work the 
children are doing. These rank orders for quality of work 
have been correlated with the rank orders for preference 
given in Table 137. The correlations are as follows : 

For gifted boys, rho =.440 
" " girls, " -.1 76 

" control boys, " =.476 

" " girls, " ^.545 

Average " =.409 

Table 138 gives for each group the per cent of those rating 
a subject who rated it above 3, indicating positive liking, 
and the per cent who rated it below 3, indicating positive 
dislike. It will be seen that in the case of both gifted girls 
and control girls, every subject is more liked than disliked. 
Gifted boys, however, dislike penmanship and folk dancing 
more than they like them, and control boys have more dis- 
like than liking for dramatics. 

This table is not as significant as it might be, for the 
reason that some groups tend on the whole to give higher 
ratings than other groups. In some respects it is more im- 
portant to know, for a given group, which subjects have 
higher per cents of 1 and 2 ratings than the average subject 
for that group, and which have higher per cents of 5 and 6 
ratings. This would tell us which subjects the group likes 
more than it likes the average school subject, and which it 
dislikes more than the average. Subjects more liked by a 
group than the average subject may be classed as positive 



SCHOLASTIC AND OTHER INTERESTS 

TABLE 138 

PER CENT OF RATINGS INDICATING POSITIVE LIKING AND 
POSITIVE DISLIKE 

(Above 3 = -f- ; below 3 = -) 



369 



Subject 


Gifted Boys 


Control Boys 


Gifted Girls 


Control Girls 




4- 


- 


+ 


- 


+ 


- 


+ 


- 


Drawing 


69.3% 


14.9% 


72.3% 


8.4% 


34.1% 


2.3% 


76.9% 


6.7% 


Modeling 


71.4% 


9.5% 


47.4% 


36.8% 


86.3% 


2.0% 


62.5% 


25.0% 


Music 


















(Instrumental) 


62.5% 


7.1% 


77.5% 


11.3% 


86.5% 


2.7% 


89.1% 


6.3% 


Music (Singing) 


55.2% 


15.6% 


65.2% 


18.8% 


76.2% 


1.7% 


85.5% 


3.6% 


Painting 


66.3% 


16.8% 


77.3% 


6.8% 


74.7% 


11.0% 


75.7% 


4.3% 


Composition 


58.3% 


11.5% 


72.3% 


13.8% 


81.7% 


4.3% 


64.1% 


16.2% 


Debating 


80.7% 


10.5% 


48.9% 


29.3% 


87.2% 


6.4% 


40.5% 


27.0% 


Dramatics 


74.4% 


9.3% 


29.4% 


41.2% 


97.9% 


0.0% 


62.5% 


31.3% 


Grammar 


62.7% 


17.6% 


61.2% 


16.5% 


75.0% 


12.0% 


68.1% 


9.9% 


Literature 


89.2% 


3.2% 


67.2% 


9.0% 


91.4% 


0.0% 


79.7% 


10.2% 


Penmanship 


29.6% 


41.7% 


62.8% 


16.8% 


51.1% 


20.2% 


72.1% 


13.5% 


Reading 


95.3% 


0.0% 


84.9% 


4.2% 


93.8% 


0.0% 


82.9% 


8.5% 


Spelling 


80.0% 


3.8% 


79.8% 


10.1% 


77.9% 


7.3% 


89.7% 


4.3% 


History 


















(Anc. or Med.) 


86.3% 


8.4% 


80.8% 


9.6% 


86.9% 


4.8% 


50.0% 


12.5% 


Civics or 


















Citizenship 


64.2% 


6.0% 


71.7% 


13.3% 


64.9% 


3.5% 


53.6% 


17.9% 


U. S. History 


92.2% 


1.7% 


85.6% 


4.8% 


79.8% 


6.4% 


63.5% 


18.8% 


Arithmetic 


77.6% 


11.2% 


69.6% 


14.4% 


73.4% 


16.0% 


78.1% 


10.5% 


Folk Dancing 


26.9% 


46.2% 


60.0% 


20.0% 


73.4% 


5.3% 


80.0% 


7.1% 


Games and 


















Sports 


96.9% 


0.0% 


95.0% 


2.0% 


93.3% 


1.1% 


79.2% 


17.9% 


Physical 


















Training 


83.3% 


3.9% 


83.0% 


11.7% 


77.7% 


4.3% 


85.0% 


5.0% 


Agriculture 


69.0% 


8.6% 


66.7% 


8.3% 


77.8% 


0.0% 


69.2% 


15.4% 


Cooking 





- 


- 


- 


89.7% 


5.7% 


78.8% 


3.0% 


Manual 


















Training 


84.5% 


6.2% 


96.2% 


1.3% 


_ 


- 


- 





Sewing 


- 


- 


- 


- 


74.2% 


14.6% 


78.3% 


12.0% 


Shop Work 


85.2% 


8.2% 


81.6% 


2.6% 


_, 


_ 





_ 


General Science 


83.7% 


2.3% 


86.1% 


8.3% 


68.2% 


4.5% 


75.0% 


2.5% 


Geography 


76.6% 


5.6% 


83.2% 


7.5% 


75.0% 


7.6% 


72.6% 


12.3% 


Nature Study 


80.8% 


2.7% 


80.4% 


5.9% 


88.5% 


6.6% 


77.5% 


7.5% 


Physiology 


















and Hygiene 


53.2% 


14.5% 


68.0% 


18.0% 


54.8% 


17.7% 


65.2% 


19.6% 



(P) for that group; those disliked more, as negative (N) ; 
those both liked an average amount and disliked an average 
amount, indifferent (I) ; those both liked and disliked more 
than average, bipolar (B). Table 139 gives this classifica- 
tion of the subjects based on the above figures. 



370 TRAITS OF GIFTED CHILDREN 

TABLE 139 
CLASSIFICATION OF THE SCHOOL SUBJECTS ACCORDING TO PREFERENCE 

Gifted Control Gifted Control 

Subject Boys Boys Girls Girls 

Drawing N I I P 

Modeling I N P N 

Music (Instrumental) I P P P 

Music (Singing) N N I P 

Painting N P N P 

Composition N N P N 

Debating P N B N 

Dramatics P N P N 

Grammar N N N I 

Literature P I P P 

Penmanship N N N N 

Reading P P P P 

Spelling P P B P 

History (Anc. or Med.) P P P N 

Civics or Citizenship I N I N 

U. S. History P P B N 

Arithmetic B N N P 

Folk Dancing N N I P 

Games and Sports P P P B 

Physical Training P P P P 

Agriculture I I P N 

Cooking P P 

Manual Training P P 

Sewing N P 

Shop Work P P 

General Science P P I P 

Geography P P N B 

Nature Study P P B P 

Physiology and Hygiene N N N N 

It will be recalled that each child was asked to mark with 
a cross each subject that was "very easy" for him, and with 
two crosses the subject that was "easiest of all." In summar- 
izing the responses it has been necessary to disregard the 
distinction between single and double crosses, as the number 
of double crosses given any one subject by our 11, 12, and 13 
year old groups was too small to justify separate treatment. 
Table 140 gives the per cent of children of each sex and 
intelligence group who marked a subject with either one or 
two crosses. 



SCHOLASTIC AND OTHER INTERESTS 



371 



TABLE 140 
PER CENTS MARKING EACH SUBJECT As "EASY" 



Subject 



Gifted 


Control 


Gifted 


Control 


Boys 


Boys 


Girls 


Girls 


22.8% 


48.7% 


28.4% 


36.5% 


4.8% 


21.1% 


11.8% 


12.5% 


28.6% 


32.4% 


39.2% 


46.9% 


19.8% 


35.7% 


42.9% 


47.3% 


15.8% 


33.0% 


22.0% 


35.7% 


29.2% 


35.1% 


50.5% 


30.8% 


31.6% 


17.1% 


38.3% 


24.3% 


32.6% 


23.5% 


55.3% 


31.3% 


29.4% 


24.3% 


53.3% 


34.1% 


31.2% 


23.9% 


58.0% 


8.5% 


9.3% 


29.2% 


28.7% 


30.6% 


80.2% 


63.9% 


77.1% 


50.4% 


56.2% 


50.5% 


70.5% 


62.4% 


43.2% 


42.3% 


54.8% 


30.0% 


20.9% 


30.0% 


33.3% 


25.0% 


46.1% 


51.9% 


50.0% 


26.0% 


48.6% 


40.8% 


50.0% 


48.2% 


11.5% 


40.0% 


50,9% 


45.0% 


38.1% 


57.0% 


51.7% 


25.5% 


28.4% 


31.9% 


46.8% 


28.8% 


24.1% 


33.3% 


33.3% 


53.8% 








57.5% 


62.1% 



Drawing 

Modeling 

Music (Instrumental) 

Singing 

Painting 

Composition 

Debating 

Dramatics 

Grammar 

Literature 

Penmanship 

Reading 

Spelling 

Ancient History 

Civics or Citizenship 

U. S. History 

Arithmetic 

Folk Dancing 

Games and Sports 

Physical Training 

Agriculture 

Cooking 

Manual Training 

Sewing 

Shop Work 

General Science 

Geography 

Nature Study 

Physiology and Hygiene 

Table 141 gives, separately for sex and intelligence, the 
ranks of the school subjects according to the per cent 
designating them as easy. The subjects at the top are those 
relatively easiest for the gifted; those at the bottom are 
relatively easiest for the control. In general, the abstract 
subjects are well toward the top, and the more "practical" 
subjects in the lower part of the lists. Surprisingly, however, 
shop work is rated relatively much easier by gifted than by 
control boys, and arithmetic somewhat easier by control 
than by gifted girls. 



26.8% 44.4% 



21.3% 53.1% 



26.2% 


13.2% 








25.6% 


50.0% 


38.6% 


45.0% 


37.4% 


47.7% 


52.2% 


22.6% 


20.5% 


45.1% 


42.6% 


32.5% 


24.2% 


30.0% 


33.9% 


30.4% 



372 



TRAITS OF GIFTED CHILDREN 



TABLE 141 
RANKS OF THE SCHOOL SUBJECTS ACCORDING TO EASE 



Subject 



Boys 

Ranks Ranks Differ- 
by by ence 
Gifted Control C-G 



Subject 



Girls 

Ranks Ranks Differ- 
by by ence 
Gifted Control C-G 



Relatively easier for G. boys 



Debating 

Dramatics 

Literature 

Grammar 

Shop Work 

Arithmetic 



9 

8 

10 

11 

16 

3 



Ancient History 5 

Little difference 
Instru. Music 13 
Physical Train. 14 
Spelling 2 

Composition 12 
Physiol. & Hyg. 18 
Civics 21 

Reading 1 

Geography 7 

U. S. History 4 
Modeling 27 

Games & Sports 6 
Agriculture 19 



26 
24 
23 
22 
27 
11 
10 



17 

18 

4 

14 

19 

20 

1 

7 

3 

25 

2 

15 



17 

16 

13 

11 

11 

8 

5 



4 
4 
2 
2 
1 
1 



-1 
-2 
-4 
-4 



Relatively easier for G. girls 

Literature 3 27 24 

Geography 8 25 17 

Ancient History 6 19 13 

Games & Sports 9 22 13 

Dramatics 5 15 10 

U.S. History 13 21 8 

Grammar 7 13 6 

Physical Train. 14 20 6 

Composition 11 16 5 

Debating 19 24 5 



Little difference 






Reading 1 


5 


4 


Civics 21 


23 


2 


Folk Dancing 10 


10 





Spelling 2 


1 


-1 


Modeling 27 


26 


-1 


Cooking 4 


2 


-2 


Nature Study 16 


14 


-2 


Physiol. & Hyg. 20 


18 


-2 


Relatively easier for C. 


girls 




Arithmetic 12 


6 


-6 


Penmanship 23 


17 


-6 


Singing 15 


7 


-8 


General Science 18 


9 


-9 


Instru. Music 17 


8 


-9 


Painting 25 


12 


-13 


Drawing 24 


11 


-13 


Agriculture 22 


3 


-19 


Sewing 26 


4 


-22 



Relatively easier for C. boys 

Penmanship 26 21 -5 

Manual Train. 15 9 -6 

Painting 24 16 -8 

Singing 23 13 -10 

General Science 17 5 -12 

Folk Dancing 25 12 -13 

Nature Study 22 8 -14 

Drawing 20 6 -14 

Table 141 yields the following rank-order correlations 
with respect to ease: 1 

Numerous studies have been made, chiefly in Germany, of children's 
interest in the different school subjects. Perhaps the most important is 
that of George Brandell, Das Interesse der Schulkinder in den Unterrichts- 
fachern; Beiheft, Zeitschrift f. ange. PsychoL, Leipzig, 1915, 168 pages. 
Others are as follows: Karl Kohn, uber Beliebtheit und Unbeliebtheit von 
Unterrichtsfachern, Zeitschrift f. pad. PsychoL, 19, 296-335, 1918; F. Malsch, 
Das Interesse f. d. Unterrichtsfachern an hoheren Knabenschulen, Zeit- 
schrift f. ange. PsychoL, 22, 393-441, 1923; W. Stern, liber Beliebtheit u. 
Unbeliebtheit der Schulfacher, Zeitschrift f, pad. PsychoL, 1905, 267-296. 
These studies allow many interesting comparisons with the material re- 
ported in the present chapter. However, as the methods which the authors 
used in collecting and treating their data are not identical with those here 
employed, any detailed comparison of results is likely to be misleading, 
and for this reason we have not attempted it. 



SCHOLASTIC AND OTHER INTERESTS 373 



Gifted boys v 
girls ' 
" boys < 
girls ' 
" boys ' 
Control boys ' 


s. control boys, rl 

" girls, 

(( 

" boys, 
gifted girls, 
control " 


w =371 
=.091 
=.013 
=.164 
=.697 
=.262 



The closest agreement (.697) is between gifted boys and 
gifted girls. Between gifted and control boys the correla- 
tion is only .371; between gifted and control girls it is too 
low to be significant (.09). 

The rank orders for the preferred and easy subject 
(Tables 139 and 141) yield the following correlations: 

Gifted boys, rho -.611 

Control " " =.638 

Gifted girls, " =.561 

Control " " =555 



Average, .591 

OCCUPATIONAL INTERESTS 

A statistical summary of the occupational interests ex- 
pressed by a group of children still in the pre-high school 
grades can have, of course, only a suggestive value. Perhaps 
no other kind of interest is so likely to change. In voicing 
such preferences the young child is doubly handicapped; he 
has little understanding of himself, and his knowledge about 
many occupations is exceedingly vague or inaccurate. Never- 
theless, even though the preferences expressed by our gifted 
children may express only ephemeral interests, they were 
deemed a necessary part of our records. They at least 
represent in each case a cross-section view of the child's 
interests, and their real value can only be appraised when 
the children have become men and women and have chosen 
their careers. 

The Interest Blank (p. 374) called for information on oc- 
cupational interests. 

The list of 125 occupations was increased to 164 by addi- 
tions which the children themselves made. These have been 
classified under twelve heads, as indicated in Table 142, 
which gives the per cent of boys and girls in each group 
naming each occupation. The first four columns give the 



374 



TRAITS OF GIFTED CHILDREN 



INTEREST BLANK 

PUT ONE CROSS BEFORE EACH OCCUPATION YOU MAY POSSIBLY DECIDE TO FOL- 
LOW. PUT TWO CROSSES BEFORE THE ONE OCCUPATION YOU ARE MOST LIKELY 
TO CHOOSE. 



.Acrobat 

Boxer or wrestler 

Auto racer 

Baseball player 

-Aviator 

.Electrician 

.Wireless operator 

Draftsman 

Mechanic or machinist 

Photographer 

Surveyor 

Clerk 

Mail carrier 

Ticket or express 

agent 

Auto salesman 

Traveling salesman 

Stenographer or typist 

Linotypist 

Private secretary 

Bookkeeper or 

accountant 
.Nurse 

Merchant 

Advertiser 

Building contractor 

Factory or business 

manager 
Banker 

Music teacher 

Singer 

Musician (player) 

Musician (composer) 

Orchestra conductor 

Dancer 

Actor or actress 

Stage manager 

Orator 

Lecturer 

Decorator 

Cartoonist 

.Magazine illustrator 

Artist (painter of 

pictures) 

Landscape artist 

Architect 

Sculptor 

If the occupation you would like best is not given above, write it here 

If you are a girl, do you prefer the duties of housewife to any other occupation? 



Section hand 

Street laborer 

Bootblack 

Logger 

Hod carrier 

Janitor 

Teamster 

Deliveryman 

Truck driver 

Waiter or waitress 

......Cook 

......Cobbler 

.Barber 

Butcher 

Baker 

Grocer 

Soldier or sailor 

Conductor or 

motonnan 

Fireman or brakeman 

Chauffeur 

Policeman 

Detective 

Plasterer 

Tailor 

House painter 

Plumber 

Carpenter 

Stone or brick mason 

Joiner 

. .Watch repairer 

Dressmaker 

Milliner 

Housewife 

. Chef 

Florist 

Fisherman 

Forest ranger 

Farmer or rancher 

Dairyman 

Stock breeder 



Library assistant 

Librarian 

Social worker 

Teacher 

School principal 

.Reporter 

Editor 

Story writer 

.Novelist 

Poet 

Play writer 

Historian 

Christian Science 

healer 

Veterinary doctor 

Dentist 

Surgeon 

Physician 

Civil engineer 

Mining engineer 

Mechanical engineer 

Electrical engineer 

Chemical engineer 

Army or navy officer 

Politician 

Congressman 

Mayor 

Lawyer or judge 

Astronomer 

Mathematician 

Physicist 

Chemist 

Mineralogist 

Botanist 

Zoologist 

Bacteriologist 

Psychologist 

College professor 

Explorer 

Priest 

Preacher 

.^..-Reformer 
......Statesman 



SCHOLASTIC AND OTHER INTERESTS 



375 



per cents for occupations the children say they are "most 
likely to follow"; the last four columns those they "may 
possibly decide to follow." 



TABLE 142 

PER CENT OF GIFTED AND CONTROL GROUPS 

OF OCCUPATIONS 



CHOOSING VARIOUS TYPES 



Type of 
occupation 


Most likely to follow 
Boys Girls 
Con- Con- 
Gifted trol Gifted trol 


May possibly follow 
Boys Girls 
Con- Con- 
Gifted trol Gifted trol 


Agriculture, 
horticulture, 
stock-raising, 
dairying 


3.8% 


0.0% 


0.2% 


0.0% 


8.2% 


7.9% 


1.7% 


0.8% 


Mechanical trades, 
building and con- 
struction and 
manufacturing 5.5% 


13.2% 


0.6% 


5.4% 


8.7% 


13.5% 


3.7% 


11.1% 


Transportation 


5.4% 


7.2% 


0.0% 


0.6% 


3.4% 


9.3% 


0.0% 


0.4% 


Commercial 


8.5% 


8.1% 


2.0% 


1.0% 


6.5% 


7.5% 


0.6% 


1.1% 


Public service 


5.1% 


10.8% 


0.0% 


0.0% 


7.3% 


10.1% 


0.3% 


0.3% 


Domestic and 
personal service 


0.5% 


2.1% 


22.8% 


15.3% 


1.1% 


2.5% 


10.5% 


13.1% 


Clerical 
occupations 


4.7% 


7.2% 


5.9% 


19.6% 


6.2% 


5.8% 


6.7% 


12.0% 


Artistic 


9.1% 


4.9% 


23.6% 


14.0% 


10.3% 


7.3% 


27.2% 


21.8% 


Semi- 
professional 


3.8% 


2.1% 


4.5% 


3.0% 


5.2% 


3.0% 


6.4% 


4.9% 


Professional 


48.3% 


24.9% 


37.8% 


37.6% 


35.1% 


20.1% 


40.5% 


31.8% 


Social work 


1.2% 


0.7% 


1.7% 


1.0% 


0.9% 


0.6% 


1.5% 


0.8% 


Athletics 


3,9% 


11.5% 


0.6% 


2.4% 


1.1% 


12.5% 


0.9% 


12.5% 



The gifted show greater preference for the following 
occupations: public service, professional (boys), artistic, 
semi-professional, and agriculture (slightly). The control 
group expresses greater preference for the following: me- 
chanical, etc., transportation, athletic, and clerical. The 
groups show little difference in preference for commercial 
occupations and social work. There are more first choices 
for domestic and personal service (including secretarial 
work) by the gifted, but more second choices by the control 
group. 



14.2 


12.8 


13.4 


11.7 


13.8 


12.9 


13.6 


12.7 


15.1 


10.9 


13.8 


12.7 


14.1 


10,6 


13.7 


12.4 


15.4 


11.8 


13.8 


13.3 


14.6 


12.0 


13.8 


12.9 


16.0 


13,0 


13.9 


12.8 


14.9 


11.4 


14.0 


12.9 


15.7 


14.0 


14.7 


12.3 


14.8 


12.8 


14.2 


12.0 


16.1 


13.9 


14.2 


12.9 


15.9 


12.7 


14.4 


12.8 



376 TRAITS OF GIFTED CHILDREN 

Table 143 gives the mean Barr Scale ratings of first 
choice and second choice occupations for each group by age 
and sex. (See page 66 & , for description of the Barr Scale of 
rating occupations according to their estimated demands 
upon intelligence.) As there were ordinarily several "second 9 * 
choices for a given child, these were averaged to give a 
single Barr rating for occupations considered as possibilities. 

TABLE 143 

MEAN BARR SCALE RATINGS OF PREFERRED OCCUPATIONS 
Most likely to follow May possibly follow 

Boys Girls Boys Girls 

Age Gifted Control Gifted Control Gifted Control Gifted Control 

8 

9 

10 
11 
12 
13 

Average all 

ages 15.4 12.7 13.9 12.6 14.7 12.1 14.0 12.6 

Average S.D. 

for all ages 2.44 1.82 1.84 1.12 

Both groups display a good deal of ambition in their oc- 
cupational preferences, but the average Barr rating of the 
occupations chosen by the gifted runs higher than that for 
the control by roughly 1.5 the S.D. of the former. The aver- 
age gifted boy is looking to an occupation which presents 
about the intellectual difficulties of high school teaching, 
preaching, or industrial chemistry; the average control boy, 
to an occupation about as intellectual as the work of a nurse, 
chef, or landscape gardener. The occupations followed by 
the fathers of the gifted have an average Barr rating of 12.77, 
as compared with 8.8 for average adult males in these cities. 
In the gifted group there is less distance between the occupa- 
tional ambitions of child and occupational status of father, 
than in the case of the control group. The occupational am- 
bitions of the control group tend to be, intelligence consid- 
ered, more extravagant than those of the gifted. 

The only significant sex difference is that the occupa- 
tional choices of gifted boys rate somewhat higher than 
those of gifted girls. 



SCHOLASTIC AND OTHER INTERESTS 377 

The total number of occupations marked by each child 
was tabulated by age, sex, and intelligence. As will be seen 
from Table 144, no significant difference was found be- 
tween the gifted and control groups. Boys of both groups 
mark about 25 per cent more occupations than girls mark, 
and for both sexes and both intelligence groups the number 
marked at age 13 is less than in earlier years. 

TABLE 144 
NUMBER OF OCCUPATIONS PREFERRED 

Average for each age 
Boys Girls 

Age Gifted Control Gifted Control 

8 7.0 7.4 5.0 6.4 

9 9.2 8.2 4.8 6.6 

10 7.0 7.0 5.3 5.1 

11 6.5 7.2 5.5 4.9 

12 6.0 8.0 7.0 5.5 

13 4.3 6.3 5.4 4.1 

Average of averages 6.67 7.3 5.5 5.4 

PREFERENCE FOR VARIOUS TYPES OF ACTIVITY 

One item in the Interest Blank was as follows : 
5. Below are several different kinds of things to do. ON THE LINE 

BEFORE EACH THING, PUT A FIGURE (1, 2, 3, 4, OR 5) TO SHOW HOW 
WELL YOU LIKE TO DO THAT KIND OF THING. ' 

Put a 1 if you LIKE IT VERY MUCH. 

Pur a 2 if you LIKE IT FAIRLY WELL. 

Put a 3 if you NEITHER LIKE IT NOR DISLIKE IT. 

Put a 4 if you RATHER DISLIKE IT, 

Put a 5 if you DISLIKE IT VERY MUCH. 

Studying your lessons. 

General reading (books, magazines, newspapers). 

Practicing music, drawing, dancing, etc. 

Playing games that require little physical exercise. 

Playing games that require lots of exercise. 

Playing with several other persons. 

Playing with one other person. 

Playing alone. 

Going to parties, picnics, dances, club meetings, etc. 

Using tools or working with apparatus and machinery. 

Sewing, cooking, knitting, housework, etc. 

Being leader in a team or club and managing other persons. 



378 TRAITS OF GIFTED CHILDREN 

The ratings were made by the child during school hours 
and with no suggestions from anyone. They may fairly safely 
be taken to represent in each case the child's attitude at the 
time the blank was filled out. The ratings were distributed 
separately by age for the boys and girls of both gifted and 
control groups. As the age differences proved to be rela- 
tively small, the ages have been combined for each group. 
Table 145 gives the mean ratings for each type of activity. 
The figures are for ages 8 to 13 only, as there were no con- 
trol children younger than 8 or gifted children older than 13. 

TABLE 145 

MEAN PREFERENCE RATINGS ON VARIOUS ACTIVITIES BY GIFTED AND 
CONTROL CHILDREN 

(1 is high and 5 is low) 

Boys Girls 

Activity Gifted Control Gifted Control 

Study lessons 2.34 2.26 2,81 1.99 

General reading 1-31 1.77 1.21 1.75 

Music, drawing, or dancing 2.84 2.81 I.b7 a./J 

Games requiring little physical exercise 2.18 2.07 2.15 2.09 

Games requiring much physical exercise 1.50 1.72 1.49 1.68 

Playing with several other persons 1.50 1.51 1.49 1.60 

Playing with one other person 1.67 2.40 1.76 2.27 

Playing alone 3.47 3.85 * 3.78 

Parties, picnics, club meetings, etc, 1.96 1.91 1.43 1.50 

Tools, apparatus, or machinery 1.81 1.89 2.97 3.47 

Sewing, cooking, housework, etc. 4.21 4.17 2.13 1.89 

Being leader in a team or club, etc. 2.12 2.20 2.21 2.37 

It will be seen that the gifted children rate most of these 
activities higher than do the control; they seem to have a 
little more enthusiasm about things in general. The control 
rate no activity appreciably higher than the gifted rate it, but 
the gifted rate reading and "playing with one other person" 
lower than do the control. The other activities are about 
equally liked by the two groups. The evidence of Table 
145 suggests rather convincingly that the interests of gifted 
children are in these respects quite normal. The typical 
gifted child likes vigorous games, plays with tools and ap- 
paratus, likes the companionship of others, and shows no 
abnormal fondness for solitude or study. The rank orders 



SCHOLASTIC AND OTHER INTERESTS 379 

of the activities for gifted and control boys give a correlation 
of .87; those for gifted and control girls, .86. 

Comparing girls with boys, we find that girls' ratings 
tend to run a little higher than those of boys. They especially 
rate "practicing music, drawing, and dancing," "going to 
parties and picnics," and "sewing, cooking, etc.," higher than 
do boys. Reading and "playing with one other person" are 
also liked somewhat better by girls. Only one activity is 
rated higher by boys, "using tools, etc." 

COLLECTIONS 

Statements of the children. In the Interest Blank, which 
was filled out by both gifted and control groups, the follow- 
ing question was asked: Name all the collections you have 
made; tell how old you were when you made the collection 
and tell how large it was. Then followed space for recording 
eight collections. The mean number of collections made is 
shown in Table 146 for age, sex, and intelligence groups. 

TABLE 146 
NUMBER OF COLLECTIONS MADE (TESTIMONY OF THE CHILDREN) 

Gifted Control 

Age Boys Girls Boys Girls 

6 1100 

7 .75 .06 

8 1.42 1.17 .48 .88 

9 1.81 1.44 .38 .97 

10 1.63 1.40 .5 .7 

11 1.68 1.93 1.15 .8 

12 1.90 1.77 1.28 1.54 

13 2.00 1.13 1.3 .74 

Ages combined, average Ages combined, average 

for both sexes 1.51 for both sexes .87 

This table is based upon total number of collections 
made, regardless of their nature. Table 147 shows that 
gifted children tend to make more collections of scientific 
value or interest. Collections of the following kinds were in- 
eluded in this category : 

Birds, birds' eggs, nests, feathers, etc. 

Coins and money 

Flowers, grasses, leaves, etc. 



380 TRAITS OF GIFTED CHILDREN 

Foreign articles, (Chinese bronzes, flags, ancient 

weapons) 

Rocks, stones, minerals, pebbles, agates, sand 
Shells, starfish, crabs, snails, insects of various sorts 
Stamps 

Electrical instruments 
Different kinds of woods. 

TABLE 147 

NUMBER WHO HAVE MADE COLLECTIONS OF SCIENTIFIC INTEREST 

OR VALUE 

Gifted Control 

Under 7 years B y s Girls B y s Girls 

Number 1400 

Average per child 0.2 0.8 

Ages 7 to 9 years 

Number 61 113 16 16 

Average per child 0.55 0.75 0.1 0.2 

Ages 10 to 13 years 

Number 130 230 76 105 

Average per child 0.68 1.11 0.30 0.57 

Ages combined 

Number 152 300 58 94 

Average per child 0.62 0.96 0.30 0.48 

Average, sexes combined 0.85 0.41 

The above tables show that 1,74 times as many gifted as 
control children have made collections of some kind, but 
that 2.07 times as many gifted as control have made collec- 
tions of scientific nature. 

Statements of the parents. In the Home Blank parents 
were asked to name the collections the child had made, to 
indicate from what age to what age it was in progress, and 
to state whether it was "large, medium, or small." Data for 
gifted only are available from the Home Blank. Replies 
were received for 603 cases of the main group ; 330 boys and 
273 girls. Failure to reply was counted as meaning that no 
collection had been made. The results are shown in 
Table 148. 



SCHOLASTIC AND OTHER INTERESTS 381 

TABLE 148 

COLLECTIONS NAMED BY PARENTS, AND AGE AT WHICH THEY 
WERE BEGUN 

Gifted Boys Gifted Girls 

Below 10- Below 10- 

7 7-9 13 Total Per cent 7 7-9 13 Total Per cent 



Stamps 13 


61 


25 


105 


31.8% 


3 


9 


6 


21 


7.7% 


Coins 3 


3 


4 


12 


3.6% 





1 





2 


0.7% 


Pictures 9 


8 


9 


29 


8.8% 


6 


5 


7 


23 


8.4% 


Shells 11 


9 


4 


25 


7.6% 


6 


4 


2 


16 


5.9% 


Insects 10 


10 


7 


28 


8.5% 


3 


4 


2 


12 


4.4% 


Stones or minerals 11 


5 


5 


25 


7.6% 


9 


4 


2 


18 


6.6% 


Flowers 4 


3 


2 


10 


3.0% 


9 


9 


5 


25 


9.2% 


Nests and eggs 4 





1 


6 


1.8% 





2 





3 


1.1% 


Electrical apparatus 


1 


2 


4 


1.2% 

















Dolls 1 











0.3% 


11 


1 


2 


31 


11.4% 


Labels and coupons 6 


6 


2 


18 


5.5% 





2 





3 


1.1% 


"Samples" 1 


2 





3 


0.9% 


7 


7 


2 


17 


6.2% 


Marbles 9 


7 





17 


5.2% 

















Postcards 5 


4 


1 


12 


3.6% 


3 


2 





5 


1.8% 


Books 13 


1 


2 


16 


4.8% 


6 


3 





15 


5.5% 


Curios 


2 





3 


0.9% 








1 


1 


0.3% 


Streetcar transfers 2 


2 


1 


5 


1.5% 

















Time tables 


3 





3 


0.9% 

















Miscellaneous 16 


12 


5 


40 


12,1% 


8 


7 


7 


29 


10.6% 


Total collections 124 


148 


82 


362 


109.7% 


76 


68 


43 


221 


80.9% 



Note. "Total" columns include a few cases for whom age was 
not stated and who therefore do not appear in the age columns. 
The "miscellaneous" category includes maps, "funny papers/* ar- 
rowheads, beads, soil, baseball records, buttons, and other collections 
receiving very infrequent mention. 

Attention is called to the fact that the ages of the 603 
subjects entering into Table 148 ranged from 3 to 13 years 
at the time the data were collected. Had all the subjects 
been 12 years old, say, the figures in the total columns would 
have been much higher, as it has been found that the collect- 
ing interest does not normally reach its maximum before 10 
or 11. Even so, the boys have averaged more than one col- 
lection each. 

The sex differences are much as might be expected. 
Boys more often than girls collect stamps, coins, marbles, 
labels, coupons, shells, insects, etc. Girls more often than 
boys collect flowers, dolls, and "samples." 



382 TRAITS OF GIFTED CHILDREN 

The following numbers of gifted children, according to 
the statements of parents, have made more than one collec- 
tion: 

Boys Girls 

N N 

Total who have made collections 188 57% 116 42.5% 

Have made two or more 92 27.9% 59 21.6% 

three " " 46 13.9% 30 11.0% 

" four " " 22 6.7% 10 3.6% 

five " " 10 3.0% 5 1.8% 

In general, the statements of the parents agree fairly well 
with those of the children themselves. Boys, however, re- 
port a good many more collections than parents report for 
them. The parent is probably more likely to forget a given 
collection or to consider it too trivial to mention. 

SUMMARY 

1. With certain exceptions, gifted children are more in- 
terested than unselected children in school subjects which 
are abstract, and less interested in the "practical" subjects. 
Their interest is relatively much stronger, for example, in 
such subjects as literature and dramatics, and much weaker 
in penmanship, manual training, sewing, etc. However, the 
gifted and control children express about the same degree 
of preference for games and sports ; also for grammar. 

2. The subject preferences of gifted boys resemble those 
of control boys far more than the preferences of gifted girls 
resemble those of control girls. The preferences of gifted 
boys and gifted girls are more alike than those of control 
boys and control girls. 

3. The average correlation of the gifted children's pref- 
erences, and the teachers' estimates of the quality of the 
children's work in the different subjects, is .41. 

4. For each sex and intelligence group, each school sub- 
ject has been compared with the average of all the subjects 
with respect to the number of 1 or 2 and 3 or 4 ratings it 
secured. This made possible a classification of the school 
subjects, for each sex and intelligence group, as "positive," 
"negative," "indifferent," or "bipolar." 

5. Subjects that are positive both with gifted boys and 
with gifted girls are dramatics, literature, reading, history, 



SCHOLASTIC AND OTHER INTERESTS 383 

games and sports, and physical training; subjects that are 
negative with both gifted groups are painting, grammar, 
penmanship, and physiology and hygiene. Civics or citizen- 
ship is the only subject that is indifferent to both groups. Bi- 
polarity is rare with both gifted and control groups. 

6. The gifted, far oftener than the control, rate as "very 
easy" such subjects as literature, grammar, debating, and 
ancient history; the control, far oftener than the gifted, 
such subjects as sewing, drawing, painting, general science, 
singing, folk dancing, penmanship, etc. However, shop work 
is ranked much higher for ease by gifted boys than by con- 
trol boys, and arithmetic somewhat higher by control girls 
than by gifted girls. 

7. Gifted girls show a considerable resemblance to gifted 
boys in respect to subjects found easy (.70), but no signifi- 
cant resemblance to control girls (.09). The sex differences 
are much greater in the control than in the gifted group. 

8. The average correlation between preference and ease 
is .59. 

9. The occupations preferred by the gifted rate higher on 
the Barr Scale than those preferred by the control, by about 
1.5 the SIX of the latter. However, there is less distance be- 
tween the mean occupational rating of parents and children 
in the case of the gifted than in the case of the control group. 

10. Of a variegated list of twelve kinds of activities rated 
by gifted and control groups, nearly all were rated higher 
by the gifted. Gifted children have more enthusiasms than 
average children, and their interests appear to be in gen- 
eral no less wholesome. 

11. One and three-quarters times as many gifted as con- 
trol children have made collections, and more than twice as 
many have made collections of scientific nature. 



CHAPTER XIV 



PLAY INTERESTS, KNOWLEDGE, AND PRACTICE 1 

Our data on play fall into two groups: (1) results of a 
questionnaire-test of interest in and knowledge of plays, 
games, and amusements; (2) replies to certain questions in 
the supplementary Home Blank and School Blank relating 
to play and to associations with other children. On most 
points comparative data were secured for a control group, 
as it was one of the main purposes of this section of the 
study to determine to what extent and in what respects the 
play life of gifted children deviates from the normal. The 
belief is generally entertained that the deviation is consider- 
able, although no statistical evidence bearing on the ques- 
tion has ever been presented. The importance of the prob- 
lem was thought to justify a considerable expenditure of 
time and effort in connection with the present investigation. 

NATURE OF THE TEST ON PLAYS, GAMES, AND AMUSEMENTS 

The purpose of this test was to secure more accurate in- 
formation about a child's play practice, play interest, and 
play knowledge than could be secured by ordinary question- 
ing. Play knowledge, of course, readily lends itself to meas- 
urement by the usual type of information test, but reliable 
data on play practice and play interests are much more dif- 
cult to secure. There are several possible methods of ap- 
proach, and it is to be regretted that there was not time for 
a comparative try-out of a number of the most promising. 
Instead, it was necessary to arrange a method for immediate 
use with but little of the preliminary experimental work 
which is so necessary for thoroughly satisfactory results in 
the field of test making. 

The test was prepared in August and September, 1921, 
by the writer, assisted by Mr. Ruch, Miss Marshall, and Miss 
Goodenough. First, the most important statistical studies 

1 Written with the assistance of Helen Marshall and Florence L. Good- 
enough. 

385 



386 TRAITS OF GIFTED CHILDREN 

of children's play activities were reviewed in order to select 
those upon which to base questions. Children in the Palo 
Alto schools were questioned as to the games they played, 
and numerous adults prepared lists of well-remembered 
games. Since it was not possible to use for this purpose more 
than a small fraction of the games which children play, the 
aim was to make a selection which would give a fair sam- 
pling of the most generally known games of all the leading 
types. Other considerations were: (1) to avoid, so far as 
possible, games which are known by many different names; 
and (2) to take account of sex and age differences in such a 
way that the test would be equally valid for boys and girls 
and over a wide range of ages. Local and seasonal charac- 
ters also had to be taken into account. Such studies as those 
of Croswell 1 and McGhee 2 afforded valuable data on the 
familiarity of various games. Types of games to be consid- 
ered included such categories as the quiet and the active, the 
social and the solitary, the competitive and the non-com- 
petitive, the intellectual and the non-intellectual, etc. Finally 
a provisional list of about 150 games, plays, and amusements 
was drawn up and arranged under the following three cate- 
gories : 

(1) Games which are in most cases fairly active and are 
usually or frequently performed alone. We may call this 
the active-solitary class. Examples are spinning tops, riding 
a bicycle, rolling hoops, fishing, and using tools. 

(2) Games which are social and usually but not always 
competitive. We may call this the social-active class. Ex- 
amples are: playing tag, hide and seek, follow the leader, 
shinny, baseball, etc, 

(3) Games which are mildly social but relatively quiet, 
making less demand upon physical strength and skill than 
upon the powers of imagination or logical thought, such as 
"playing" school, dominoes, cards, charades, checkers, etc. 

The provisional list was submitted for criticism to a num- 
ber of individuals, was then reduced to 120, and was then 
tried out on about a hundred children of grades four, six, 

*T. R. Croswell : Amusements of Two Thousand Worcester School Chil- 
dren. Pedagogical Seminary, Vol. 6, pp. 314-371. 

3 Z. McGhee: A Study in the Play Life of Some South Carolina Chil- 
dren. Pedagogical Seminary, Vol. 7, pp. 459-478. 



PLAY INTERESTS 387 

and eight. On the basis of the data so collected the list was 
reduced to 90. For use in the test the plays were listed in 
three columns corresponding to the types active-solitary, 
active-social, and social-quiet. The words in each column 
are so arranged that there is an age progression from the 
top to the bottom, those near the top referring to activities 
which appeal to young children and those near the bottom 
to the older. Although the order was based upon subjective 
judgments and is far from accurate, it was borne out in a 
general way by the data later collected from unselected 
children. 

The entire test was made up in an eight-page booklet, 
8 l /2 x 11, containing seven exercises. Three of these were de- 
voted to the above-mentioned 90 games and amusements. 
The fourth called for testimony regarding experience and 
skills, and the last three constituted a test of play informa- 
tion. The material is reproduced in the following pages : 



388 



TRAITS OF GIFTED CHILDREN 



EXERCISE 1. (INSTRUCTIONS) 
First, put ONE CROSS, X, on the line before each thing you HAVE EVER DONE 

OR PLAYED. 
Next, put TWO CROSSES, XX, before each thing you CAN DO OR PLAY VERY 

WELL. 



BEGIN HERE. 
....Roll hoops 
....Play jackstones 
....Spin tops 
....Play jackstraws 
....Walk on stilts 

....Fly kites 

....Play with bow and arrow 

....Coast or toboggan 

....Ride bicycle 

....Skate 

....Ski 

....Hike 

....Do garden work 

....Dance 

....Shoot 

....Fish 
....Swim 

Ride horseback 

Row a boat 

....Hunt 

Do plain sewing 

....Cook a meal 
Knit, crochet or do fancy- 
work 

....Use tools 
....Work with machinery 



Play tag 

....Ring around the rosy 
....London Bridge 
....Farmer in the dell 
... Jn and out the window 

....Hide and seek 

....Hopscotch 

....Drop the handkerchief 

....Blindfold 

....Postofflce 

....Cat and mouse 
....Red Rover 
....Puss in the corner 
....Pom-pom pull-away 
....Blackman 

....Marbles 
....Duck on rock 
....Follow the leader 
....Anty over 

....Dare base or prisoner's 
base 

....Snap the whip 
....Tug of war 
....Roly poly 
....Jump the rope 
....Leapfrog 

....Fox and hounds 
....Fox and geese 
....Shinny 
....Croquet 
....Bowling 

....Wrestling 
....Baseball 
....Racing or jumping 

Handball 

Soccer 

....Boxing 

....Tennis 

....Volleyball 

....Basketball 

....Football 



EXERCISE 2. (INSTRUCTIONS) 

Put ONE CROSS before each thing you LIKE TO DO OR PLAY. 
Put TWO CROSSES before each thing you LIKE VERY WELL. 
(The word-lists were repeated.) 



....Play with dolls 
....Play "dress up" 
....Play house 
....Play store 
....Play school 

....Simon says thumbs up 
....Play church 
....Dominoes 
....Crokinole 
....Parchesi 

....Tiddledy-winks 
....Backgammon 

Authors 

....Snap 
....Guessing games 

....Cards 
....History cards 
....Geography cards 
....Charades 

...Anagrams or word build- 
ing 

....Solve puzzles 

....Checkers 

....Chess 

....Pool 

....Billiards 



PLAY INTERESTS 389 

EXERCISE 3. (INSTRUCTIONS) 

Put ONE CROSS before each thing you do or play ONCE OR TWICE A MONTH. 
Put TWO CROSSES before each thing you do or play ONCE OR TWICE A WEEK. 
Put THREE CROSSES before each thing you do or play NEARLY EVERY DAY. 
(The word-lists were again repeated.) 

EXERCISE 4 
Read each question and draw a line under the right answer. 

1. Did you ever catch a fish? YES NO 

2. Did you ever shoot any game? YES NO 

3. Did you ever make a trap? YES NO 

4. Have you ever made a bow and arrow? YES NO 

5. Have you ever made a kite? YES NO 

6. Have you ever set up an electrical apparatus? YES NO 

7. Did you ever make a water wheel? YES NO 

8. Have you ever made a canoe or boat that one could ride in? YES NO 

9. Can you row a boat? YES NO 

10. Have you ever learned the wireless code? YES NO 

11. Can you stand on your head? YES NO 

12. Have you ever climbed a tree? YES NO 

13. Did you ever swim 100 feet? YES NO 

14. Can you dive? YES NO 

15. Can you chin yourself? (Chin a pole) YES NO 

16. Can you walk on you hands? YES NO 

17. Can you turn a handspring? YES NO 

18. Are you a Boy Scout? YES NO 

19. Are you a first-class Boy Scout? YES NO 

20. Are you a Bluebird? YES NO 

21. Are you a Campfire Girl? YES NO 

22. Are you a Girls* Reserve? YES NO 

23. Are you a Girl Scout? YES NO 

24. Can you read the time from a sundial? , YES NO 

25. Have you ever set up a tent? YES NO 

26. Have you ever hiked eight miles a day? _ -YES NO 

27. Have you ever cooked a meal? , YES NO 

28. Have you ever made a dress that was good enough to wear? YES NO 

29. Have you ever had a paper route or sold papers? YES NO 

30. Have you ever milked a cow? YES NO 

31. Have you ever hitched up a horse? YES NO 

32. Have you ever driven a horse? YES NO 

33. Have you ever 'driven a tractor? -YES NO 

34. Have you ever been elected to any office or special honor? YES NO 

35. Have you ever been captain of an athletic team? YES NO 

36. Do you belong to a football team? YES NO 

37. Do you belong to a baseball team? YES NO 

38. Are you a member of a debating team? YES NO 

39. Do you belong to a hiking club? YES NO 

40. Do you belong to a sewing club? - YES NO 

41. Do you belong to a track team? YES NO 

42. Do you belong to a gardening or stock-raising club? YES NO 

43. Do you belong to any kind of musical or dramatic club? YES NO 

44. Do you belong to any kind of church society or church club? YES NO 

45. Have you ever taken part in a play? YES NO 



390 TRAITS OF GIFTED CHILDREN 

Moat of the questions in Exercise 4 relate to activities of 
the kind which normal, healthy-minded boys or girls are 
likely, under favorable circumstances, to have experience 
with whether they are very intelligent or not. If gifted chil- 
dren are typically bookish, non-active, and non-social, they 
should be expected to make a low score on this test. The 
boy who scores high on it would be, presumably, the type 
one often hears described as "real boy" healthy-minded, 
active, and socially adaptable. It is of course not assumed 
that the test is an accurate measure of these traits, but it was 
thought that it would at least yield data of suggestive value 
when a large gifted group was compared with a large control 
group, 

EXERCISES 5, 6, AND 7 

These were all of the ordinary type of information test 
with alternative response words, the correct word in each 
sentence to be underlined. The first series (19 items) is de- 
signed to test chiefly knowledge regarding the solitary-active 
plays listed under Exercises 1, 2, and 3; the second series 
(82 items) is a knowledge test regarding social-active plays; 
the third series (22 items) is a test of knowledge regarding 
the semi-social-quiet plays. In all there were 123 items. 
One purpose of Exercises 5, 6, and 7 was to serve as a check 
upon the accuracy of responses in Exercises 1, 2, and 3. 

EXERCISES 5, 6, AND 7 
Series 1. 

1. A floor for skating is called a GRIDIRON LINKS RINK. 

2. You pick up jackstraws with a MAGNET HOOK FINGERS. 

3. Stirrups are used in HOCKEY SWIMMING RIDING. 

4. A good wood for making bows to shoot arrows is CEDAR HICKORY WILLOW, 

5. "Dog fashion" is a term used in DIVING ROWING SWIMMING. 

6. Darts are most often made from SHINGLES TIN TWIGS. 

7. Skis are popular in AUSTRALIA BRAZIL CANADA. 

8. A quiver is a case for carrying ARROWS CARTRIDGES PISTOLS. 

9. A toboggan is a kind of FISHING TACKLE GUN SLED. 

10. A brand of motorcycle is the BUICK INDIAN RANGER. 

11. The Mead Bicycle Co. manufacture the COLUMBIA CRUS/DER HARLEY-DAVIDSON. 

12. A much-prized game fish is the CARP SARDINE TROUT. 

13. A reel is used in ATHLETICS PISHING HUNTING. 

14. Canvasbacks are a kind of CANOE GAME-BIRD TENT. 

15. A pommel is a part of a BRIDLE GUN SADDLE. 

16. Decoys are used in DIVING HUNTING POLO. 

17. "Blue rocks'* are used in COASTING FISHING SHOOTING. 

18. The "jack-knife" is a kind of DANCE DIVE RACE. 

19. Trolling is a term used in FISHING GOLF HUNTING. 



PLAY INTERESTS 391 

Series 2. 

1. A singing game is FARMER-IN-THE-DELL I-SPY POISON. 

2. A game often played in the school room is CAT-AND-MOUSE CROQUET BLACKMAN. 

3. You must dodge quickly in GOING-TO-JERUSALEM TAG THREE-DEEP. 

4. One of the players is blindfolded in BLACKMAN HOPSCOTCH HIDE-AND-SEEK. 

5. The word "lag" is used in playing CROKINOLE MARBLES SHINNY. 

6. A game played in the snow is FOX-AND-GEESE JACKSTRAWS POTATO-RACE. 

7. In croquet we use MALLETS NETS RACQUETS. 

8. An example of a hiding game is DARE-BASE RED ROVER RUN-SHEEP-RUN. 

9. A game of "dares** is ANTY-OVER FOLLOW-THE-LEADER TAG. 

10. One must run fast in playing CHEESE-IT POSTOFFICE THREE-DEEP. 

11. The players form in a ring in BEAR-IN-THE-PIT BLACK-TOM ROLY-POLY. 

12. Three-old-cat is most like BASEBALL BOWLING SHINNY. 

13. A kissing game is FRUIT-BASKET POSTOFFICE RING-AROUND-THE-ROSY. 

14. The Boy Scouts* motto is "BE HELPFUL" "BE PREPARED" "BE TRUTHFUL." 

15. One of the "red** honors in scouting is BIRD-NAMING HEROISM TRAILING. 

16. In the Morse code one dot is the letter A M E. 

17. The second law of the Campfire Girls is "BE HONEST" "BE HAPPY" "GIVE 

SERVICE.** 

18. A magazine of the Campflre Girls is "THE CLUB WORKER" "THE RALLY" 

"WOHELO." 

19. The number of players on a baseball team is 9 11 13. 

20. The number of balls needed to "walk" a player is 3 4 5. 

21. A drop kick scores 123. 

22. The number of players on a basketball team is 5 7 9. 

23. The record pole vault is about 11 FEET 13 FEET 15 FEET. 

24. Football is played on COURTS GRIDIRONS LINKS. 

25. The fastest runner can go a hundred yards in 8:3 9:3 10:3. 

26. Ty Cobb is a BOXER PLAYER WRESTLER. 

27. The Brooklyn Nationals are called the GIANTS INDIANS SUPERBAS. 

28. De Palma is noted for AVIATION MOTOR-RACING YACHT-RACING. 

29. The score in tennis is tied at DEUCE LovE-30 VANTAGE-IN. 

30. A dance that goes with jazz music is the FOX-TROT TWO-STEP WALTZ. 

31. The game of quoits is most like BOWLING HORSESHOES SOCCER. 

32. Hockey is sometimes played on ICE STILTS WATER. 

33. Billiards is played on a FLOOR LAWN TABLE. 

34. Roly-poly is played with a MALLET NET RUBBER BALL. 

35. Billiards is played with CARDS CUES DICE. 

36. Caddies are used in EUCHRE GOLF VOLLEY-BALL. 

37. The players form in a line in PASS-BALL SLAP-JACK TAG. 

38. Pucks are used in ARCHERY HOCKEY TENNIS. 

39. "Throwing up the sponge" is a term from BOXING HORSE-RACING WRESTLING. 

40. In Derby races the horses HURDLE PACE TROT. 

41. One must run fast in playing JACKSTONES TIN-TIN WOOD-TAG. 

42. "Migs" are used in BASEBALL MARBLES MUMBLETY-PEG. 

43. The players form in a ring in BLACK TOM OLD WITCH LONDON BRIDGE. 

44. The players form in a ring in DROP-THE-HANDKERCHIEF FOX-AND-GEESE TUG-OF- 

WAR 

45. A singing game is FOLLOW-THE-LEADER LONDON BRIDGE POISON. 

46. A game where you look for something hidden is I-SPY OLD WITCH ROLY-POLY. 

47. A game requiring chairs is FLYING DUTCHMAN GOING-TO- JERUSALEM LONDON 

BRIDGE. 

48. Wickets are used in playing CROQUET HOCKEY SOCCER. 

49. Prisoner's base is most like ANTY-OVER POM-POM-PULL-AWAY TENNIS. 

50. You must throw straight in DUCK-ON-ROCK LONDON BRIDGE RED ROVER. 

51. Shinny is played with BATS CLUBS RACQUETS. 

52. A game with forfeits is DARE-BASE FOLLOW-THE-LEADER THREE-DEEP. 



392 TRAITS OF GIFTED CHILDREN 

Series 2 (continued). 

53. The players form a ring in playing BLACKMAN THREE-DEEP TUG-OP-WAR. 

54. A game in which you must not smile is FRUIT-BASKET OLD-WITCH TIN-TIN. 

55. One of the players is blindfolded in KEEP-MOVING RUTH-AND-JACOB TUG-OF- 

WAR. 

56. An example of a counting game is Buz CRAMBO TURN-ABOUT. 

57. The leader of a Boy Scout troop is called the ADJUTANT CORPORAL SCOUTMASTER. 

58. A Boy Scout is not expected to CARRY MATCHES EAT CANDY SMOKE. 

59. A Boy Scout prides himself most on his FIGHTING HONOR KNOWLEDGE. 

60. The leader of Campflre Girls is called the GAMP-MISTRESS GUARDIAN TORCH- 

BEARER. 

61. The^ first law of the Campfire Girls is "BE HAPPY" "BE HONEST" "SEEK BEAUTY" 

62. The second degree of a Campflre Girl is FIRE-MAKER GUARDIAN TORCH-BEARER. 

63. The quarterback is a player in BASEBALL BASKETBALL FOOTBALL. 

64. The number of strikes needed to "fan" a player is 3 4 5. 

65. The regular number of innings in baseball is 9 1 13. 

66. The number of players on a football team is 7 9 11. 

67. A touchdown scores 356. 

68. The mile record for running is nearest 4 MINUTES 5 MINUTES 6 MINUTES. 

69. The record for the running broad jump is nearest 17 FEET 20 FEET 24 FEET. 

70. Dribble is a term from BASEBALL BASKETBALL HOCKEY. 

71. The Chicago Nationals are called CARDINALS CUBS WHITE Sox. 

72. A famous authority on card games was BOYLE WALTER CAMP HOYLE. 

73. Barney Oldfleld is noted for AVIATION MOTOR-RACING YACHT-RACING. 

74. Hurdles are used in GOLF POLO RACES. 

75. Soccer is most like FOOTBALL HANDBALL TENNIS. 

76. An indoor game is SOCCER TENPINS VOLLEY-BALL. 

77. Quoits are played with KNIVES MARBLES RINGS. 

78. Cockswain is a term from AVIATION HORSEMANSHIP ROWING. 

79. Spare is a term used in BILLIARDS BOWLING POOL. 

80. Putters are a kind of CLUBS DOGS PIGEONS. 

81. Horses are used in HOCKEY LACROSSE POLO. 

82. Tenpins is a kind of BOWLING BILLIARDS HOCKEY. 

Series 3. 

1. A game in which you have to add quickly is DOMINOES FRUIT-BASKET OLD 

WITCH. 

2. The "king row" is used in CHECKERS CROQUET DOMINOES. 

3. Snap is played with CARDS DICE MALLETS. 

4. The joker is used in CARDS CHECKERS TENNIS. 

5. A good parlor game is AUTHORS POTATO RACE BLACKMAN. 

6. The ace is used in CARDS CHECKERS TENNIS. 

7. The player snaps a small, flat disk in BACKGAMMON CROKINOLE PARCHESI. 

8. "Trumps** is a term used in CARDS CROKINOLE POOL. 

9. Parchesi is played with CARDS DICE DOMINOES. 

10. Whist is played with BALLS CARDS PINS. 

11. Charades is a GAME OF CHANCE GUESSING GAME KISSING GAME. 

12. "HEARTS" is played with CARDS DICE DOMINOES. r 

13. A game in which you make up words is ANAGRAMS Buz CRAMBO. 

14. Cards dealt to center of table are called ROYAL FLUSH SLIPPERY ANN THE WIDOW. 

15. Tiddledy-winks is most like BACKGAMMON CROKINOLE PAHCHESI. 

16. Checkmate is a term used in CHECKERS CHESS DOMINOES. 

17. The number of cards in a deck is 48 52 64. 

18. Pawns are used in CHECKERS CHESS CROKINOLE. 

19. The highest number of flinch cards is 10 15 25. 

20. The game of checkers is most like CHESS DOMINOES WHIST. 

21. Backgammon is played with CARDS CLUBS DICE. 

22. The number of squares on a chess board is 36 48 64. 



PLAY INTERESTS 393 

The test was given to children in groups of 20 to 50. The 
rule for timing was to allow as nearly as possible all the 
children to complete each exercise, and in no case to proceed 
to another until at least 90 per cent were through with the 
preceding one. Only a few children failed to complete all 
the tasks. As the test presupposed considerable literacy it 
was not given to subjects below the third school grade. Even 
for the third and fourth grades it is rather difficult reading. 

GROUPS TESTED 

In all about 1,200 gifted children were given the test. Of 
the 643 children of the main gifted group, 554 were tested. 
In addition, a control group was tested consisting of 474 un- 
selected children of grades 3 to 9 in Sunnyvale, Mountain 
View, and San Jose, California/ The first two cities have a 
population of less than 5,000, while San Jose has about 40,000. 
The schools chosen were known to be about average with 
respect to the intelligence and social status of the children. 
This group will be designated Control Group B. 

This chapter will be devoted entirely to the results for the 
main gifted group and for the control group. 

Table 149 gives the distribution of control and gifted 
groups by age and sex. It should be noted that the control 
group averages considerably older than the gifted group, a 
fact which needs to be taken into account in certain com- 
parisons made further on. 

TABLE 149 
SUBJECTS GIVEN PLAY TEST 



Age 


6 and 7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Total 


Control boys 





9 


19 


30 


32 


29 


42 


34 


30 


225 


Control girls 





21 


27 


23 


35 


25 


46 


48 


24 


249 


Total, control 





30 


46 


53 


67 


54 


88 


82 


54 


474 


Gifted boys 


17 


41 


48 


71 


68 


40 


18 








303 


Gifted girls 


20 


32 


43 


46 


51 


42 


17 








251 



Total, gifted 37 73 91 117 119 82 35 554 
DERIVATION OF "PREFERENCE INDICES" FOR THE 90 ACTIVITIES 

It will be recalled that Exercises 1, 2, and 3 deal, respec- 
tively, with knowledge of, interest in, and time devoted to 
the 90 different activities. The responses on these three ex- 
ercises were tabulated and summarized by age and sex for 



394 TRAITS OF GIFTED CHILDREN 

the gifted and control groups. The original intention had 
been to work out age and sex tendencies for each of the 
three exercises, but this was not found to be feasible. After 
considerable experimentation it was decided to combine the 
results of the three exercises for a single rating, as they are 
supplementary rather than independent parts of the test. 
For example, with respect to play knowledge, one might 
have supposed that the extent of familiarity with the ac- 
tivities would be definitely affected by maturity; that the 
younger child wopild report but few plays as known, and 
that older children would add more and more. This was not 
the case. Instead of increasing with age, the plays marked 
as known tend to remain of about the same number, but 
they change their character almost entirely. Thus, the 
younger children report knowledge of such activities as 
rolling hoops, playing London Bridge, playing with dolls, 
playing house and store, etc., while older children mark 
these plays very infrequently. Instead, they report activities 
common to older children, such as working with tools, play- 
ing baseball, tennis, authors, etc. This omission on the part 
of the older children robs Exercise 1 (report of play knowl- 
edge) of much of its significance. The children have tended 
to limit this part of their report to the same plays and ac- 
tivities that they mark under play interest and play prac- 
tice (Exercises 2 and 3). The result is that Exercise 1 has 
little statistical value in its present form. It is probable, 
however, that it has a certain value as a practice exercise, 
introducing the subject to the method of rating, and ac- 
quainting him with the list of activities. Accordingly, Exer- 
cise 1 was disregarded entirely in scoring, and the reports 
on Exercises 2 and 3 (play interest and play practice) were 
combined on the assumption that the composite score on the 
two exercises would give a better index of play preference 
than could be derived from either exercise taken by itself. 
Various methods of scoring Exercises 2 and 3 were con- 
sidered. The method finally adopted was as follows : a play 
marked with one cross was scored 1; one marked with two 
crosses, 2; and (in Exercise 3) one marked with three 
crosses, 4. This weighting was based upon the relative fre- 
quency of responses by one cross, two crosses, and three 
crosses. The end to be attained was a preference rating for 



PLAY INTERESTS 



395 



each of the 90 activities, First, the responses (single, double, 
and triple crosses) given by each subject in Exercises 2 and 3 
were transmuted into point scores, by the weightings given 
above, and these were tabulated, separately by activity, for 
all the children of a given group, age and sex (e. g., 10 year 
old gifted boys) . Next, the total of the numerical scores re- 
ceived by a given activity from a given age and sex group 
was found and divided by the number of children in the 
group, to give a preference score of that activity for that 
group. The method may be illustrated by the use of the fol- 
lowing hypothetical responses of five children (A, B, C, D, 
E) for five activities. 





Exercise 2. Play Interest 
Child 


Exercises. Play Practice 
Child 


Total Score 
for the Five 
Subjects 


Total Score 

-7-5 = 

Preference 
Index of 
Activity 


Activity 


A 


B 





D 


E 


A 


B 





D 


E 


13 

6 
4 
9 
6 


2.6 
1.2 
0.8 
1.8 
1.2 


Snap the whip 
Tug of war 
Roly-poly 
Jump the rope 
Leapfrog 


X 
XX 

X 


X 
XX 

X 


X 
XX 


XX 


X 
XX 


X 
XXX 


XX 
X 

XX 


XX 

X 


XXX 
X 


XX 
XX 



In this way a preference index was computed for each of 
the 90 activities, separately for each age, each sex, and each 
intelligence group (control and gifted). Also a mean pref- 
erence index for each activity was calculated for each sex of 
the control group as a whole, ages combined, and similarly 
for each sex of the gifted group, ages combined. Table 
150 gives these preference indices for the 90 activities. It 
will be noted that each game has at each age a preference 
index for control boys, one for gifted boys, one for control 
girls, and one for gifted girls. The mean preference index 
for all the ages of a group are given in the last column. 
These are derived by averaging the age means of the other 
columns, The groups are designated as follows : 

CB=Control Boys 
GB=Gif ted Boys 
CG^Control Girls 
GG-Gif ted Girls. 



396 TRAITS OF GIFTED CHILDREN 

TABLE 150 

PREFERENCE INDICES OF 90 ACTIVITIES FOR GIFTED AND 
CONTROL CHILDREN 

























Mean 














Ages 










for 






6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 




CB 





25 


21 


13 


10 


9 


6 


10 


5 


13 


Roll 


GB 


11 


8 


7 


10 


7 


4 


3 








7 


hoops 


CG 





10 


14 


9 


7 


10 


5 


4 


3 


8 




GG 


12 


9 


8 


6 


8 


4 


1 








7 




CB 





6 


7 


4 


3 


1 


1 


3 


4 


4 


Play 


GB 


11 


2 


1 


2 


1 


1 


1 








3 


jackstones 


CG 





16 


21 


9 


10 


10 


5 


5 


3 


10 




GG 


21 


14 


9 


12 


13 


12 


8 








13 




CB 





20 


18 


25 


15 


15 


20 


18 


12 


18 


Spin 


GB 


18 


13 


15 


10 


16 


13 


8 








13 


tops 


CG 





6 


10 


8 


7 


6 


3 


3 


2 


6 




GG 


10 


5 


6 


6 


5 


6 


2 








6 




CB 





3 


14 


7 


6 


5 


3 


1 


3 


5 


Play 


GB 


11 


8 


8 


4 


5 


2 











6 


jacks traws 


CG 





7 


11 


10 


9 


4 


2 


2 


2 


6 




GG 


10 


5 


5 


4 


7 


4 


6 








6 




CB 





18 


21 


28 


21 


20 


21 


19 


9 


20 


Walk on 


GB 


16 


9 


10 


11 


12 


9 


18 








12 


stilts 


CG 





14 


19 


11 


16 


17 


12 


9 


5 


13 




GG 


8 


8 


9 


10 


10 


9 


2 








8 




CB 





20 


33 


27 


24 


25 


25 


23 


18 


24 


Fly 


GB 


20 


15 


16 


16 


19 


11 


12 








16 


kites 


CG 





11 


19 


10 . 


15 


8 


10 


5 


3 


10 




GG 


12 


8 


7 


7 


6 


6 


2 








7 




CB 





17 


24 


14 


13 


10 


11 


13 


9 


14 


Play with 


GB 


12 


19 


14 


13 


13 


9 


9 








13 


bow and arrow 


CG 





9 


8 


10 


5 


6 


1 


2 


3 


6 




GG 


6 


5 


7 


6 


8 


6 


2 








6 




CB 





1 


5 


2 


5 


2 


6 


3 


4 


4 


Coast or 


GB 


11 


5 


9 


8 


6 


5 


13 








8 


toboggan 


CG 





3 


4 


3 


2 


2 


1 


2 


2 


2 




GG 


5 


2 


3 


3 


3 


6 


2 








3 




CB 





21 


30 


33 


35 


33 


43 


47 


45 


36 


Ride 


GB 


19 


24 


23 


32 


35 


30 


35 








, 28 


bicycle 


CG 





13 


16 


17 


33 


30 


27 


25 


17 


22 




GG 


14 


8 


8 


9 


19 


14 


12 








12 




CB 





23 


20 


23 


18 


20 


15 


14 


13 


18 


Skate 


GB 


18 


15 


24 


13 


18 


17 


13 






17 




CG 





20 


31 


27 


31 


29 


21 


19 


15 


24 




GG 


38 


30 


26 


26 


28 


22 


16 






27 




CB 





1 


3 


2 


4 


1 


2 


1 


1 


2 


Ski 


GB 


9 


3 


2 


1 


2 


2 


4 






3 




CG 





4 


2 


2 


1 


1 








3 


2 




GG 


1 











1 


1 













PLAY INTERESTS 



397 



TABLE 150 Continued 

























Mean 














Ages 










for 






6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 




CB 





8 


15 


12 


15 


17 


15 


17 


13 


14 


Hike 


GB 


19 


14 


18 


19 


21 


19 


23 








19 




CG 





7 


12 


6 


10 


12 


12 


18 


15 


11 




GG 


23 


11 


15 


14 


17 


20 


16 








16 




CB 





28 


22 


24 


19 


27 


23 


18 


28 


24 


Do garden 


GB 


17 


11 


14 


13 


13 


9 


7 








12 


work 


CG 





20 


21 


21 


21 


16 


11 


11 


17 


17 




GG 


30 


10 


14 


12 


10 


15 


8 








14 




CB 





3 


8 


3 


5 


6 


8 


4 


7 


6 


Dance 


GB 


9 


6 


6 


5 


5 


4 


11 








7 




CG 





29 


25 


26 


26 


27 


17 


22 


27 


25 




GG 


34 


19 


28 


26 


23 


26 


32 








27 




CB 





14 


16 


15 


19 


14 


29 


25 


22 


19 


Shoot 


GB 


14 


14 


7 


13 


15 


15 


11 








13 




CG 





13 


4 


5 


4 


3 


1 


2 


1 


4 




GG 


8 








1 


3 


1 


1 








2 




CB 





13 


18 


13 


14 


13 


19 


18 


17 


16 


Fish 


GB 


16 


12 


15 


16 


16 


13 


17 








15 




CG 





9 


11 


8 


6 


6 


4 


4 


3 


6 




GG 


15 


4 


4 


8 


9 


9 


5 








8 




CB 





11 


14 


13 


13 


14 


16 


19 


20 


15 


Swim 


GB 


12 


11 


13 


16 


19 


18 


26 








16 




CG 





10 


11 


10 


12 


8 


7 


11 


12 


10 




GG 


18 


9 


13 


16 


17 


19 


18 








16 




CB 





7 


17 


15 


17 


19 


22 


23 


19 


17 


Ride 


GB 


12 


12 


13 


12 


12 


10 


9 








11 


horseback 


CG 





13 


15 


17 


14 


12 


9 


16 


19 


14 




GG 


17 


7 


12 


10 


10 


8 


10 








10 




CB 





9 


12 


7 


12 


10 


15 


9 


14 


11 


Row a 


GB 


9 


11 


11 


14 


16 


9 


17 








12 


boat 


CG 





10 


6 


7 


3 


7 


3 


7 


5 


6 




GG 


5 


5 


4 


7 


7 


5 


9 








6 




CB 





2 


16 


14 


19 


11 


22 


20 


19 


15 


Hunt 


GB 


13 


8 


6 


7 


7 


7 


6 








8 




CG 





6 


4 


5 


3 


2 


1 


3 


2 


3 




GG 


7 

















1 








1 




CB 





3 


5 


5 


4 


2 


3 


4 


3 


4 


Do plain 


GB 


7 


3 


1 


1 


3 


1 











2 


sewing 


CG 





16 


27 


33 


23 


24 


21 


23 


37 


26 




GG 


24 


18 


24 


20 


17 


16 


12 








19 




CB 





2 


6 


10 


5 


12 


10 


9 


8 


8 


Cook a 


GB 


9 


5 


6 


5 


8 


4 


6 








6 


meal 


CG 





14 


20 


21 


23 


22 


23 


26 


40 


24 




GG 


10 


3 


9 


12 


13 


19 


16 








12 


Knit, 


CB 








5 


3 


3 


3 


1 


1 


2 


2 


crochet 


GB 


8 


2 


2 


1 


1 














2 


or do fancy 


CG 





4 


10 


15 


13 


15 


17 


14 


26 


14 


work 


GG 


15 


10 


14 


10 


12 


11 


5 








11 



398 



TRAITS OF GIFTED CHILDREN 



TABLE 150 Continued 

























Mean 














Ages 










for 






6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 




CB 




19 


19 


25 


24 


28 


33 


23 


26 


25 


Use 
tools 


GB 
CG 


19 


22 
4 


25 

4 


25 

7 


25 
3 


23 
1 


32 
1 


2 


3 


24 
3 




GG 


17 


3 


2 


6 


6 


3 


1 








5 


Work 
with 
machinery 


CB 
GB 
CG 
GG 


15 
4 


6 
11 
3 
1 


12 
10 
3 



6 
12 
5 
2 


12 
14 
2 
1 


13 
9 

1 


16 
17 
2 
1 


18 
1 


19 
1 


13 
13 
2 
1 




CB 




31 


36 


30 


20 


26 


19 


17 


11 


24 


Play 
tag 


GB 
CG 


28 


19 
30 


22 

43 


22 
33 


19 
33 


10 
31 


11 
24 


16 


13 


19 
28 


*,**& 


GG 


42 


32 


30 


27 


24 


21 


12 


. 





27 


Ring 


CB 





13 


12 


12 


8 


6 


4 


4 


5 


8 


around 
the rosy 


GB 
CG 


12 


5 
25 


3 
25 


1 
26 


1 
16 


1 
15 



5 


4 


7 


3 
15 




GG 


27 


14 


10 


6 


5 


2 


2 











CB 


... 


12 


12 


13 


9 


6 


3 


3 


5 


9 


London 
Bridge 


GB 
CG 
GG 


15 
31 


7 
26 
24 


5 

22 
20 


2 
21 
10 


2 

20 
8 


1 

16 
4 



6 
6 


5 


8 


5 

16 
15 


Farmer 


CB 


. 


18 


19 


16 


9 


9 


4 


2 


4 


10 


in the 


GB 


18 


5 


4 


2 


2 








. . 







dell 


CG 




30 


25 


25 


21 


17 


8 


6 


8 


18 




GG 


31 


23 


16 


12 


8 


5 


5 





p 


14 


In and 


CB 





7 


14 


9 


6 


4 


3 


2 


2 


6 


out the 
window 


GB 
CG 


12 


4 
18 


2 
17 


1 
18 


1 

17 



13 



8 


5 


7 


3 
13 




GG 


22 


16 


11 


8 


5 


3 


1 








9 




CB 





24 


26 


30 


19 


27 


20 


17 


9 


22 


Hide and 
seek 


GB 
CG 


32 


14 
31 


13 
39 


17 

37 


15 
27 


10 
24 


8 
18 


15 


13 


16 
26 




GG 


41 


27 


30 


23 


22 


17 


24 








26 




CB 





9 


17 


19 


9 


12 


6 


6 


5 


10 


Hop- 
scotch 


GB 
CG 


20 


8 
32 


6 
36 


5 
39 


5 

35 


2 

29 


1 
16 


8 


12 


7 
26 




GG 


33 


26 


25 


20 


18 


14 


6 


- 





20 




CB 





17 


17 


14 


11 


10 


6 


5 


7 


11 


Drop the 
handkerchief 


GB 
CG 


16 


6 
22 


5 
26 


5 
27 


2 

19 


2 

16 


2 

10 


5 


11 


5 

17 




GG 


30 


22 


13 


9 


8 


5 


2 


- 





13 




CB 


_ 


6 


11 


13 


7 


9 


4 


4 


3 


7 


Blind- 


GB 


17 


6 


5 


2 


4 


2 


1 








5 


fold 


CG 




19 


18 


20 


11 


12 


7 


5 


7 


12 




GG 


22 


16 


15 


10 


9 


6 


3 


- 





12 




CB 





11 


12 


9 


9 


8 


10 


7 


4 


9 


Post- 


GB 


18 


5 


3 


3 


4 


3 


7 


- 





6 


office 


CG 





15 


12 


9 


13 


13 


8 


8 


8 


11 




GG 


17 


6 


6 


4 


7 


5 


2 








7 



PLAY INTERESTS 



399 



TABLE 150 Continued 

























Mean 














Ages 










for 






6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 




CB 





23 


20 


12 


8 


5 


4 


1 


2 


9 


Cat and 


GB 


14 


6 


3 


4 


1 














4 


mouse 


CG 





32 


29 


26 


17 


12 


7 


4 


5 


17 




GG 


18 


13 


11 


7 


3 


2 











8 




CB 





3 


12 


8 


10 


5 


3 


2 


3 


6 


Red 


GB 


6 


4 


1 


3 


2 














2 


Rover 


CG 





9 


6 


6 


5 


4 


3 


1 


1 


5 




GG 


11 


2 


5 


2 


1 


1 


1 








3 




CB 


- 


10 


19 


13 


9 


9 


7 


8 


4 


10 


Puss in 


GB 


17 


9 


7 


6 


3 


2 











6 


the corner 


CG 





20 


22 


24 


14 


16 


7 


4 


6 


14 




GG 


26 


18 


19 


11 


7 


4 


1 








12 




CB 





9 


14 


12 


8 


4 


5 


5 


4 


8 


Pom-pom 


GB 


9 


5 


4 


6 


7 


6 


7 








6 


pull-away 


CG 





8 


10 


8 


9 


14 


5 


3 


3 


8 




GG 


10 


9 


5 


4 


6 


3 


2 








6 




CB 


.. ^ 


7 


9 


9 


6 


5 


5 


6 


3 


6 


Black- 


GB 


8 


2 


1 


2 





2 


1 








2 


man 


CG 





9 


9 


9 


8 


9 


2 


2 


6 


7 




GG 


9 


2 


1 


1 


1 


1 











2 




CB 





19 


24 


27 


24 


22 


25 


20 


11 


22 


Marbles 


GB 


19 


24 


23 


21 


22 


15 


12 








19 




CG 





4 


10 


10 


9 


8 


6 


4 


1 


7 




GG 


13 


4 


6 


6 


5 


5 


2 








6 




CB 





3 


7 


10 


7 


1 


5 


6 


6 


6 


Duck on 


GB 


11 


5 


4 


6 


2 


4 


4 








5 


rock 


CG 





6 


6 


5 


5 


3 


2 


4 


3 


4 




GG 


8 





1 





1 


2 











2 




CB 


, 


19 


21 


16 


14 


11 


10 


9 


9 


14 


Follow the 


GB 


22 


11 


15 


11 


10 


7 


10 








12 


leader 


CG 





17 


26 


23 


16 


13 


8 


7 


5 


14 




GG 


27 


15 


10 


11 


12 


8 


3 








12 




CB 





8 


14 


14 


11 


13 


11 


8 


7 


11 


Anty 


GB 


9 


3 


1 


5 


5 


6 


2 








4 


over 


CG 





7 


14 


11 


19 


8 


8 


8 


5 


10 




GG 


5 


4 


1 


5 


4 


5 


6 








4 




CB 





6 


11 


12 


10 


12 


6 


6 


6 


9 


Dare base or 


GB 


11 


8 


8 


8 


6 


6 


7 








8 


prisoner's base 


CG 
GG 


8 


5 
1 


9 
4 


18 
2 


23 
11 


20 
6 


13 
4 


1 


10 


12 
5 




CB 


_ _ 


21 


23 


17 


14 


13 


14 


11 


9 


15 


Snap 
the whip 


GB 
CG 


11 


10 
8 


10 
15 


12 
15 


9 
13 


8 
14 


9 
6 


6 


6 


10 
10 




GG 


11 


9 


7 


8 


9 


6 


2 








7 




CB 


^ 


26 


22 


21 


16 


9 


16 


13 


9 


17 


Tug of 


GB 


18 


12 


13 


14 


14 


10 


11 








13 


war 


CG 





10 


13 


15 


9 


9 


4 


2 


3 


8 




GG 


13 


7 


7 


8 


7 


6 


2 








7 



400 



TRAITS OF GIFTED CHILDREN 



TABLE 150 Continued 

























Mean 














Ages 










for 






6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 




CB 





6 


8 


4 


6 


2 


2 


1 


3 


4 


Roly- 


GB 


8 


2 





1 





2 











2 


poly 


CG 





9 


7 


4 


1 


6 











3 




GG 


11 


1 


2 


2 





3 











3 




CB 





23 


23 


19 


16 


12 


11 


7 


7 


15 


Jump 


GB 


12 


5 


6 


5 


5 


4 


3 








6 


the rope 


CG 





33 


20 


30 


31 


33 


20 


2 


18 


23 




GG 


38 


27 


25 


21 


18 


16 


9 








22 




CB 





10 


18 


19 


10 


12 


12 


9 


8 


12 


Leap- 


GB 


12 


12 


9 


12 


7 


5 


8 








9 


frog 


CG 





16 


17 


16 


10 


4 


3 


5 


3 


9 




GG 


18 


13 


10 


5 


7 


4 


1 








8 




CB 





7 


5 


9 


7 


4 


5 


4 


3 


5 


Fox and 


GB 


9 


6 


5 


4 


4 


2 


3 








5 


hounds 


CG 





8 


6 


6 


2 


4 





17 


1 


6 




GG 


6 


2 


1 


1 


1 


1 


2 








2 




CB 





17 


14 


14 


11 


13 


11 


10 


6 


12 


Fox and 


GB 


11 


4 


4 


4 


3 


2 


3 








5 


geese 


CG 





19 


14 


15 


15 


12 


6 


5 


8 


12 




GG 


12 


6 


3 


2 


1 


1 











4 




CB 





6 


6 


11 


8 


8 


8 


7 


5 


8 


Shinny 


GB 


7 


8 


11 


6 


4 


4 


3 








6 




CG 





5 


4 


3 


5 


3 


2 








3 




GG 


6 


1 


2 


2 


2 


2 











2 




CB 








5 


7 


6 


5 


6 


4 


2 


5 


Croquet 


GB 


9 


5 


7 


7 


10 


7 


6 








7 




CG 





2 


4 


4 


4 


6 


2 


4 


3 


4 




GG 


10 


5 


7 


11 


7 


10 


12 








9 




CB 





1 


6 


6 


4 


2 


4 


4 


4 


4 


Bowling 


GB 


8 


6 


2 


5 


3 


2 


5 








4 




CG 





4 


3 


5 


2 


1 











2 




GG 


3 


1 





1 


1 


1 


1 








1 




CB 





7 


21 


19 


16 


14 


26 


18 


-17 


17 


Wrestling 


GB 


12 


20 


20 


20 


17 


14 


18 








17 




CG 





3 


8 


9 


4 


3 


2 


2 





4 




GG 


6 


1 


3 


3 


7 


2 











3 




CB 





37 


34 


38 


41 


37 


43 


41 


41 


39 


Baseball 


GB 


22 


25 


31 


36 


37 


39 


33 








32 




CG 





11 


20 


29 


26 


25 


35 


30 


31 


26 




GG 


14 


10 


9 


17 


16 


24 


15 








15 




CB 





19 


19 


21 


24 


21 


18 


16 


19 


20 


Racing or 


GB 


18 


19 


19 


21 


21 


19 


21 








20 


jumping 


CG 





18 


19 


22 


20 


18 


11 


13 


10 


16 




GG 


19 


14 


13 


16 


15 


12 


8 








14 




CB 





7 


17 


12 


15 


10 


12 


7 


9 


11 


Handball 


GB 


16 


14 


22 


27 


28 


31 


27 








24 




CG 





9 


11 


16 


8 


10 


5 


6 


6 


9 




GG 


12 


9 


5 


10 


7 


18 


10 








10 



PLAY INTERESTS 
TABLE 150 Continued 



401 

























Mean 














Ages 










for 






6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 




CB 





1 


5 


8 


13 


7 


11 


10 


6 


8 


Soccer 


GB 


8 


5 


7 


13 


12 


10 


11 






10 




CG 





3 


2 


4 


1 


1 











1 




GG 


2 


1 


1 


2 


4 


2 


2 






2 




CB 





12 


11 


22 


17 


14 


23 


16 


14 


16 


Boxing 


GB 


14 


15 


10 


16 


12 


12 


13 






13 




CG 





4 


1 


. 9 


1 


2 





1 


2 


3 




CB 


_ 


2 


12 


9 


10 


10 




11 


10 


11 


2 
9 


Tennis 


GB 


13 


10 


12 


10 


14 


17 


16 






13 




CG 





6 


4 


14 


7 


10 


9 


19 


11 


10 




GG 


10 


5 


4 


11 


10 


11 


16 






9 




CB 





20 


23 


20 


20 


19 


20 


10 


10 


18 


Volleyball 


GB 


11 


5 


4 


8 


II 


7 


6 






8 




CG 





6 


11 


23 


21 


26 


17 


12 


22 


17 




GG 


8 


8 


8 


6 


11 


15 


13 






10 




CB 





12 


17 


15 


23 


18 


25 


19 


27 


20 


Basketball 


GB 


9 


7 


8 


14 


,15 


11 


20 






12 




CG 





7 


10 


13 


8 


15 


12 


17 


18 


13 




GG 


5 


6 


4 


4 


9 


14 


12 







8 




CB 





10 


20 


15 


27 


17 


20 


16 


18 


18 


Football 


GB 


12 


16 


14 


19 


18 


14 


24 






17 




CG 





4 


6 


12 


4 


5 


2 


3 


4 


5 




GG 


5 


1 


3 


1 


4 


1 









2 




CB 





6 


5 


4 


3 





1 


2 


3 


3 


Playing 


GB 


7 


1 





















1 


with dolls 


CG 





43 


39 


35 


26 


24 


13 


5 


9 


24 




GG 


43 


33 


33 


20 


12 


9 


5 







22 




CB 


. 


6 


5 


3 


3 


3 


1 


1 


1 


3 


Play 


GB 


5 


1 


3 


1 





1 










2 


"dress up" 


CG 





33 


34 


34 


24 


21 


11 


7 


8 


22 




GG 


36 


34 


29 


20 


14 


9 


6 








21 




CB 


: 


19 


9 


8 


5 


1 


2 


3 


2 


6 


Play house 


GB 


9 


2 


2 



















2 




CG 





37 


36 


33 


23 


20 


10 


6 


5 


21 




GG 


38 


32 


25 


18 


11 


7 











19 




CB 





16 


12 


8 


6 


5 


2 


3 


2 


7 


Play store 


GB 


12 


4 


4 


2 


2 














3 




CG 





29 


23 


25 


17 


19 


9 


6 


4 


17 




GG 


33 


24 


21 


12 


11 


5 











15 




CB 





17 


16 


8 


7 


3 


2 


4 


2 


8 


Play school 


GB 


13 


5 


4 


2 


2 


1 











4 




CG 





36 


33 


27 


22 


21 


13 


11 


9 


21 




GG 


36 


30 


22 


11 


13 


7 


2 








17 




CB 





10 


10 


11 


9 


6 


6 


3 


2 


7 


Simon says 


GB 


16 


6 


10 


5 


4 


2 


1 








6 


thumbs up 


CG 





15 


14 


18 


18 


10 


7 


4 


3 


11 




GG 


19 


16 


15 


7 


7 


5 


3 








10 



402 



TRAITS OF GIFTED CHILDREN 



TABLE 150 Continued 

























Mean 














Ages 










for 






6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 




CB 


. 


3 


8 


3 


4 


1 


1 


6 


1 


4 


Play 
church 


GB 
CG 


6 


1 
18 


1 
17 



13 




7 



8 



1 


1 


3 


1 
8 




GG 


16 


4 


5 


1 


1 


1 











4 




CB 





10 


13 


13 


8 


13 


7 


8 


6 


10 


Dominoes 


GB 
CG 


14 


12 

8 


15 
14 


14 
14 


14 
9 


9 
14 


10 

7 


7 


5 


13 
10 




GG 


16 


13 


12 


13 


12 


12 


6 








12 




GB 








5 


5 


4 





2 





1 


2 


Grokinole 


GB 


9 


1 


4 


3 


9 


5 


7 








5 




CG 
GG 


4 


2 

1 


2 

1 


4 
3 


4 


2 


1 








2 




CB 





8 


6 


4 


6 


8 


2 





2 


4 


Parchesi 


GB 
CG 


11 


6 
3 


11 
5 


10 
3 


10 
2 


7 
5 


6 
2 


5 


3 


9 

4 




GG 


9 


9 


7 


12 


11 


14 


8 








10 




CB 


. 


11 


13 


11 


6 


5 


3 


4 


1 


7 


Tiddledy- 
winks 


GB 
CG 


11 


6 
10 


9 
10 


8 
11 


6 
9 


5 

10 


2 
3 


2 


3 


7 
7 




GG 


13 


10 


8 


9 


7 


8 


2 








8 




CB 








4 


4 


3 





1 


1 


1 


2 


Back- 


GB 


5 


1 


1 


1 


2 





1 








1 


gammon 


CG 





2 


3 


3 


1 














1 




GG 


3 


1 




























CB 


._ . 


2 


6 


6 


3 


2 


1 


2 


1 


3 


Authors 


GB 


14 


4 


6 


5 


8 


6 


5 








7 




CG 





3 


3 


7 


4 


4 


3 


4 


3 


4 




GG 


9 


6 


5 


9 


9 


9 


5 








9 




CB 


.. 


10 


9 


7 


6 


2 


3 


2 


2 


5 


Snap 


GB 


9 


2 


4 


1 


3 


1 


2 








3 




CG 





7 


11 


7 


4 


4 


2 


3 


2 


5 




, GG 


6 


2 


1 


3 


2 


2 


2 








3 




* CB 





9 


6 


10 


9 


8 


5 


7 


3 


7 


Guessing 


GB 


15 


6 


12 


8 


8 


7 


2 








8 


games 


CG 





11 


13 


23 


17 


14 


12 


9 


12 


14 




GG 


23 


14 


9 


20 


15 


17 


11 








15 




CB 


- 


20 


21 


16 


16 


23 


13 


17 


9 


17 


Cards 


GB 


23 


13 


20 


19 


19 


15 


17 





- 


18 




CG 





11 


16 


19 


16 


17 


15 


17 


18 


16 




GG 


28 


19 


16 


18 


22 


20 


16 








20 




CB 





1 


4 


7 


7 


6 


3 


1 


3 


4 


History 


GB 


6 


4 


5 


4 


5 


3 











4 


cards 


CG 





3 


5 


7 


5 


3 


1 


3 


7 


4 




GG 


7 


2 


2 


4 


5 














3 




CB 








6 


5 


6 


5 


4 


3 


3 


4 


Geography 


GB 


6 


3 


5 


4 


6 


3 


2 








4 


cards 


CG 





5 


3 


8 


3 


2 


2 


2 


6 


4 




GG 


7 


2 


4 


2 


2 


1 


2 








3 



PLAY INTERESTS 



403 



TABLE 150 Concluded 























Mean 












Ages 










for 




6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 


Group 


CB 








4 


4 


3 





1 


1 


1 


2 


GB 


6 


1 


3 


2 


2 


1 


1 








2 


CG 





4 


2 


3 


1 


2 


1 


2 





2 


GG 


5 





1 


7 


5 


8 


9 








5 


CB 





6 


8 


8 


6 


5 


4 


4 


3 


5 


GB 


6 


3 


5 


2 


4 


2 











3 


CG 





6 


3 


10 


5 


10 


3 


2 


2 


5 


GG 


11 


2 


1 


2 


2 


3 


2 








3 


CB 








6 


8 


10 


12 


9 


9 


3 


7 


GB 


14 


11 


16 


13 


16 


8 


18 








14 


CG 





5 


9 


15 


11 


15 


11 


10 


10 


11 


GG 


12 


14 


17 


13 


15 


15 


19 








15 


CB 





8 


18 


17 


13 


17 


17 


16 


15 


15 


GB 


24 


15 


20 


20 


23 


18 


23 








20 


CG 





13 


11 


17 


10 


14 


11 


13 


15 


13 


GG 


25 


14 


15 


14 


17 


17 


14 








16 


CB 





1 


6 


4 


4 


1 


3 


1 


2 


3 


GB 


6 


5 


4 


3 


9 


7 


11 








6 


CG 




3 


1 


5 


1 


2 


1 


1 


1 


2 


GG 


6 


1 


2 


1 


2 


2 











2 


CB 





10 


11 


6 


8 


3 


6 


7 


9 


7 


GB 


6 


5 


3 


5 


6 


3 


7 








5 


CG 




6 


2 


2 


1 


2 





1 





2 


GG 


7 


2 





1 


1 


3 











2 


CB 





4 


3 


4 


8 


2 


3 


2 


7 


4 


GB 


9 


5 


4 


5 


4 


2 


6 


. 





5 


CG 




4 


1 


3 





2 


1 








2 


GG 


6 











1 


2 











1 



Charades 



Anagrams 
or word 
building 



Solve 
puzzles 



Checkers 



Chess 



Pool 



Billiards 



It was thought that a comparison of these preference in- 
dices for the gifted and control groups might show signifi- 
cant influence of the factor of intelligence in determining 
play interests. It is seen from Table 150, however, that 
these were determined chiefly by sex and age and only to a 
slight extent by intelligence. This is shown by the following 
correlations between the mean preference indices of the 
ninety games for the boys and girls of the two groups (using 
the figures of the last column in Table 150) : 

Pearson r 

Gifted boys vs. control boys .83 .022 

Gifted boys vs. gifted girls -20 .068 

Gifted boys vs. control girls .18 .069 

Control boys vs. control girls .35 .062 

Control boys vs. gifted girls .22 .068 

Gifted girls vs. control girls -82 .022 

That is, the correlation between two opposite-sex groups is 



404 TRAITS OF GIFTED CHILDREN 

always low, even in the same intelligence class; that between 
two same-sex groups is always high, even if the intelligence 
class is different. These correlations offer no support to the 
popular belief that the gifted boy is effeminate in his play 
interests. In fact, the correlation between gifted boys and 
either gifted or control girls (.20 and .18) is lower than that 
between control boys and either of the feminine groups 
(.35 and .22). 

The preference indices of Table 150, although they 
make possible interesting sex comparisons, are not, as they 
stand, satisfactory preference indices of the various activi- 
ties. To a degree they are invalidated by the fact that the 
figures for the four groups (CB, GB, CG, and GG) are not di- 
rectly comparable, since they are affected by the tendency 
of some of the groups to mark more or fewer activities than 
do other groups. This defect can be remedied by transmut- 

oc 
ing the raw preference indices of each group into values. 

For example, the mean preference indices for the CB group 
(last column of Table 150) were distributed, the o was cal- 

/y 

culated, and each of the 90 indices was given its value. 

G 

This was done for each of the groups CB, GB, CG, and GG, 
giving four sets of values, as shown in Table 151. 

(T 

TABUE 151 

x 
PREFERENCE INDICES OF ACTIVITIES EXPRESSED IN VALUES 

Series 1. CB GB CG GG 

Roll hoops +0.28 -0.29 -0.36 -0.34 

Play jackstones -0.96 -0.89 -0.08 +0.53 

Spin tops +0.96 +0.61 -0.64 -0.49 

Play jackstraws -0.82 -0.44 -0.64 -0.49 

Walk on stilts +1.24 +0.46 +0.35 -0.20 

Fly kites +1.78 +1.06 -0.08 -0.34 

Play with bow and arrow +0.41 +0.61 -0.64 -0.49 

Coast or toboggan -0.96 -0.14 -1.21 -0.92 

Ride bicycle +3.43 +2.86 +1.62 +0.38 

Skate +0.96 +1.21 +1.90 +2.56 

Ski -0.23 -0.89 -1.21 -1.36 

Hike +0.41 +1.51 +0.06 +0.97 

Do garden work +1.78 +0.46 +0.91 +0.68 

Dance -0.68 -0.29 +0.63 +2.56 

Shoot +1.10 +0.61 -0.93 -1.07 



PLAY INTERESTS 405 
TABLE 151 Continued 

Series 1 (continued). CB GB CGr GG 

Fish +0.69 +0.91 -0.64 -0.20 

Swim +0.55 +1.06 -0.08 +0.97 

Ride horseback +0.82 +0.31 +0.49 +0.09 

Row a boat 0.00 +0.46 -0.64 -0.49 

Hunt +0.55 -0.14 -0.07 -1.21 

Do plain sewing -0.54 -1.04 +0.77 +1.40 

Cook a meal -0.41 -0.44 +1.90 +0.38 

Knit, crochet or do fancy work -1.23 -1.04 +0.49 +0.24 

Use tools +1.92 +2.26 -1.07 -0.63 

Work with machinery +0.28 +0.61 -1.21 -1.21 

Series 2. 

Play tag +1.78 +1.51 +2.47 +2.56 

Ring around the rosy -0.41 -0.89 +0.63 -0.05 

London Bridge -0.27 -0.59 +0.77 +0.82 

Farmer in the dell -0.13 -0.59 +1.05 +0.68 

In and out the window -0.68 -0.89 +0.35 -0.05 

Hide and seek +1.51 +1.06 +2.19 +2.42 

Hopscotch -0.13 -0.29 +2.19 +1.55 

Drop the handkerchief 0.00 -0.59 +0.91 +0.53 

Blindfold -0.54 -0.59 +0.20 +0.38 

Postoffice -0.27 -0.44 +0.06 -0.34 

Cat and mouse -0.27 -0.74 +0.91 -0.20 

Red Rover -0.68 -1.04 -0.79 -0.92 

Puss in the corner -0.13 -0.44 +0.49 +0.38 

Pom-pom pull-away -0.41 -0.44 -0.36 -0.49 

Blackman -0.68 -1.04 -0.50 -1.07 

Marbles +1.51 +1.51 -0.50 -0.49 

Duck on rock -0.68 -0.59 -0.93 -1.07 

Follow the leader +0.41 +0.46 +0.49 +0.38 

Antyover 0.00 -0.74 -0.08 -0.78 

Dare base or prisoner's base -0.27 -0.14 +0.20 -0.63 

Snap the whip +0.55 +0.16 -0.08 -0.34 

Tug of war +0.82 +0.61 -0.36 -0.34 

Roly-poly -0.96 -1.04 -1.07 -0.92 

Jump the rope +0.55 -0.44 +1.76 +1.84 

Leapfrog +0.14 +0.01 -0.22 -0.20 

Fox and hounds -0.82 -0.59 -0.64 -1.07 

Fox and geese +0.14 -0.59 +0.20 -0.78 

Shinny -0.41 -0.44 -1.07 -1.07 

Croquet -0.82 -0.29 -0.93 -0.05 

Bowling -0.96 -0.74 -1.21 -1.21 

Wrestling +0.82 +1.21 -0.93 -0.92 

Baseball +3.84 +3.46 +2.19 +0.82 

Racing or jumping +1.23 +1.66 +0,77 +0.68 

Handball 0.00 +2.26 -0.22 +0.09 

Soccer -0.41 +0.16 -1.35 -1.07 



408 TRAITS OF GIFTED CHILDREN 

activities. These are the masculinity indices sought. This 

oc 
method gave plus values for activities more preferred 

by boys than by girls, and minus values for activities more 
preferred by girls. To avoid the use of negative quantities, 
equivalent numerical indices were computed, on a scale of 

cc 
1 to 25, with 13 corresponding to zero values of , 1 being 

x 
the lowest negative value of and 25 the highest positive 

value. This gave the masculinity indices of Table 152. 

From Table 152 the order of the activities from most to 
least masculine is found to be as follows : 

24. Tools. 

23. 

22. 

21. Shooting. 

20. Kites, bicycle, marbles, wrestling, boxing, football. 

19. Tops, machinery, baseball. 

18. Fishing. 

17. Bow and arrow, skiing, tug of war, soccer. 

16. Stilts, garden work, basketball, pool. 

15. Hoops, swimming, rowing, hunting, snap the whip, shinny, rac- 
ing and jumping. 

14. Coasting, hiking, riding, duck on rock, leapfrog, bowling, hand- 
ball, backgammon, checkers, chess, billiards. 

13. (Line of neutrality.) Red Rover, pom-pom pull-away, follow the 
leader, anty over, roly-poly, fox and geese, croquet, volleyball, 
dominoes, crokinole, parchesi, tiddledy-winks, snap, cards, his- 
tory cards, geography cards, word building. 

12. Jackstraws, postoffice, blackman, fox and hounds, tennis, authors, 
11. Tag, hide and seek, puss in corner, dare base, Simon says, play- 
ing church, solving puzzles. 

10. Jackstones, skating, drop the handkerchief, blindfold, 
9. Ring around the rosy, -London Bridge, farmer in the dell, in and 
out the window, cat and mouse, jumping rope, guessing games, 
charades. 

8. Dancing, sewing, playing store. 
7. Knitting or crocheting. 
6. Playing school. 
5. Cooking, playing house. 
4. Hopscotch. 
3. Dressing up. 
2. Dolls. 



PLAY INTERESTS 



409 



TABLE 152 

MASCULINITY INDICES OF 90 ACTIVITIES BASED ON SEX DIFFERENCES 
FOUND IN THE CONTROL GROUP 

(Above 13 = more liked by boys; below 13, more liked by girls) 



15 Roll hoops 

10 Play jackstones 

19 Spin tops 

12 Play jackstraws 

16 Walk on stilts 

20 Fly kites 

17 Bow and arrow 
14 Coast or toboggan 

20 Ride bicycle 
10 Skate 

17 Ski 

14 Hike 

16 Do garden work 
8 Dance 

21 Shoot 

18 Fish 

15 Swim 

14 Ride horseback 

15 Row a boat 
15 Hunt 



11 Play tag 
9 Ring around rosy 
9 London Bridge 
9 Farmer in the dell 
9 In and out window 

11 Hide and seek 
4 Hopscotch 

10 Drop handkerchief 
10 Blindfold 

12 Postoffice 



2 Play with dolls 

3 Play "dress up" 

5 Play house 
8 Play store 

6 Play school 

11 Simon says thumbs up 
11 Play church 
13 Dominoes 
13 Crokinole 
13 Parchesi 



9 Cat and mouse 13 Tiddledy- winks 
13 Red Rover 14 Backgammon 

11 Puss in corner 12 Authors 
13 Pom-pom pull-away 13 Snap 

12 Blackman 9 Guessing games 



20 Marbles 

14 Duck on rock 

13 Follow the leader 

13 Anty over 

11 Dare base 



8 Do plain sewing 15 Snap the whip 

5 Cook a meal 17 Tug of war 

7 Knit or crochet 13 Roly-poly 

24 Use tools 9 Jump rope 

19 Work with machinery 14 Leapfrog 

12 Fox and hounds 

13 Fox and geese 
15 Shinny 

13 Croquet 

14 Bowling 

20 Wrestling 

19 Baseball 

15 Racing^or jumping 
14 Handball 

17 Soccer 

20 Boxing 

12 Tennis 

13 Volleyball 

16 Basketball 
20 Football 



13 Cards 
13 History cards 
13 Geography cards 
9 Charades 

13 Word building 

11 Solve puzzles 

14 Checkers 
14 Chess 

16 Pool 

14 Billiards 



410 



TRAITS OF GIFTED CHILDREN 



MASCULINITY RATINGS OF THE CHILDREN 

The masculinity indices of the activities were not derived 
primarily for the purpose of comparing the activities them- 
selves, but to serve as the basis of masculinity ratings of the 
individual children. It is obvious that when the activities 
have been rated for masculinity it will be possible so to rate 
the child, on the basis of the activity preferences which he 
expresses. The method of accomplishing this may be illus- 
trated by the following actual responses of a child: 



Activity 

Roll hoops 

Skate 

Hike 

Shoot 

Row a boat 
Dare base 

Baseball 

Handball 

Soccer 

Boxing 

Volleyball 

Basketball 

Football 
Parchesi 
Checkers 



Responses Responses 
Given in Given in 
Exercise 2 Exercise 3 

X 
X 



Total Point 
Score Value 

of 

Responses 

in Exercises 

2 and 3. 



Mascu- Mascu- 
linity Unity 
Index of Score of 
the Ac- the 
tivity. Responses 
(See for Each 



(Seep. 395) Table 152) Activity 



X 

XX 
X 

XX 

XX 

X 

XX 
X 
X 



XXX 



XXX 
XX 



X 

X 



X 
X 



X 
X 
X 

X 
X 
X 



15 
10 
14 

21 
15 
11 



15 
10 
14 

21 

90 
11 



1 X 

1 X 

2 X 

32 



X 


19 


= 114 


X 


14 


56 


X 


17 


17 


X 


20 


60 


X 


13 


13 


X 


16 


32 



20 
13 
14 



20 
13 
28 



Total point value of responses 
Masculinity score of responses 514 

Dividing 514 by 32 gives a masculinity rating of 16 for this child. 

A masculinity rating was computed for each child of the 
gifted and control groups. The distributions of these ratings, 
with means and S.D's are given in Tables 153 and 154. 



PLAY INTERESTS 



411 



CQ 

c/T 

> 
o 

CQ 



O 



c .. 

3 i 



CO 

O 
55 



3 



s^ 



o 



o 



O 







l CO CO CO ZO (?<l C<J OO CO C<J- 

CO l>- Ci CO 



TH T 1 Tt< -**! TJH CT) Ttl T-H T-H TH 

T^ CO ^* CO CO 



?q CO T-l CD TF rH 



CO C^ *> C* 



C<1 <^5 CO T^ Od T 1 



CM ^ XO 1C -^f 



C^ CO 1O t^ OO l>- 



CO CQ O-Cft OO 



UO 



CO G^l <O OO UO CO 



rH CM C<l CM t- C<l i-H 



-tf JO 
CO T-H 






&$ 

Oa CO CD 



412 



TRAITS OF GIFTED CHILDREN 



1 






O 

Q 






I a 







3 co s 
8 *r* 
gob 



TH XO CO CO Xt> O T 



05 xo 10 cq <o -^ 



cq <o - 



T-t IO CO "tf 1 1 



1J OO OO -^ 



TH rH Tt< CO CD <^ ^ 



TH TH QC OO t- (M T 



c>ai t^ 10 CD so TH 



CO 1O 

<rq^ 







rHlTS 



15 



OO CO 

"d? c^ 



^H 
^ 



TH O TH 
CMTH 



PLAY INTERESTS 413 

TABLE 154 

SIGNIFICANCE OF THE MASCULINITY DIFFERENCES FOUND IN 
TABLES 152 AND 153 



Control and Gifted Boys 


Control and Gifted Girls 




Diff. 


o 


Diff. 


Diff. 


<y 


Diff. 




between 


of the 


G 


between 


of the 


G 


Age 


means 


diff. 


diff. 


means 


diff. 


diff. 


8 


1.28 


0.385 


3.32 


-0.20 


0.393 


-0.51 


9 


0.47 


0.383 


1.23 


-0.22 


0.335 


-0.66 


10 


1.11 


0.319 


3.48 


+0.15 


0.367 


-0.41 


11 


0.48 


0.294 


1.63 


+0.62 


0.314 


+1.97 


12 


0.56 


0.321 


1.74 


+0.75 


0.296 


+2.53 


13 


-0.20 


0.272 


-0.74 


+0.32 


0.332 


+0.96 



Total 0.32 0.145 2.21 +0.12 0.154 +0.78 

It will be seen that the difference is as great as twice the 
a of the difference only at the following ages : for boys, at 8 
and 11 ; for girls, at 11 and 12. 

Figure 25 gives the age curves for mean masculinity 
ratings of gifted and control groups by sex. 

It will be recalled that in the establishment of the mas- 
culinity index, 13 was the value arbitrarily assigned to the 
game of neutral masculinity, with a range of from 1 to 25. 
This fact must be kept in mind in the interpretation of 
Tables 152 and 153 and Figure 25. 

A comparison of the curve for gifted boys with that for 
control boys confirms the conclusions on page 404 regarding 
the normal masculinity of the bright boy. The mean mas- 
culinity index for gifted boys is definitely higher than that of 
control boys at each age until 13, where it is practically the 
same as for the control boys. In the case of gifted girls the 
curve is somewhat irregular, but its general tendency is to 
run as high as that for the control girls. One striking fact 
about the curves of mean masculinity ratings of the control 
girls is the noticeable falling off in masculinity which occurs 
after the age of 14. The gifted girls show a slight tendency in 
this direction after age 12. 

Table 155 gives the proportion of gifted at each age in 
various percentile ranges of the control group. 

MATURITY INDICES OF ACTIVITIES 

From Table 150, showing the amount of preference ex- 
pressed by the control children of each age for each ac- 



414 TRAITS OF GIFTED CHILDREN 

tivity, it is possible to compute for each activity a maturity 
index. First, for a given activity curves were plotted for the 
control boys and control girls, showing for each sex the 

FIGURE 25 
MEAN MASCULINITY RATINGS BY AGE, GIFTED AND CONTROL GROUPS 



BOYS 



Neutra^ L/r?e> 




11 

\ _ ^ 

GIRLS 



6-7 8 9 10 11 \Z 13 14 15-17 



amount of increase or decrease of preference by age. By 
means of mechanical smoothings the tendency of a curve 
was then expressed in a straight line. The significant factor 
is the size of the angle, positive or negative, included be- 
tween this line and a horizontal line projected from its point 
of departure. An activity showing decreasing popularity 
with age would give an angle in the fourth quadrant, a nega- 
tive angle; one showing increasing popularity with age, 



PLAY INTERESTS 



415 



1 

o 



oo co 

CO CO 

oo co 



o 
o 



S S 



o 
cq 



o 
o 



tf 

w 



O TH 

S 
& 



tO 4O 

TH O 



^ TH OS O5 

O5 lO OO OO 



10 
tO 



w 

I 



CO 

o 



S 2 

o * 



oo oq 
OS i> 



o 
o 



CO 

co 



cq 
to* 



00 OO 



rj< co 

00 OO 



CO tO to OO 
CO ^ O C* 

TH 10 CJI 



to oo 

CO TH 



o 

o 

CO 

g 
S 



00 OO 



o 
o 



TH O 

00 



00 ^t< 

co crS 



TH O CO 

oo to co* 
to c<i 



J 
o 



a 



* 



OJ to 
<M 



o ^ 

o oo 

O TH 



Oi iO 
TH OJ 



O 
O 



fc 
o 

g 
S 




*j O 

w CN 



g 



O ^ 

cvi to 



is I a 



416 



TRAITS OF GIFTED CHILDREN 



an angle in the first quadrant, or positive angle. Since the 
size of the angle is affected by the scale on which the curve 
is plotted, the tangents of the angles have been compared, 
instead of the angles themselves. Finally, all tangents were 
expressed as multiples of the tangent of an angle of 6% 
which is approximately an even decimal (.1). Thus, the 
ratio of a given angle's tangent to the tangent of an angle 
of 6 becomes the maturity index of that activity. These 
ratios ranged from +5.95 for bicycle riding (control boys), 
showing a very rapid increase of interest with age, to 
11.34 for doll play (gifted girls), showing a very rapid de- 
crease of interest. Table 156 gives the maturity indices of 
all the activities by age, sex, and intelligence. 

TABLE 156 
MATURITY INDICES OF ACTIVITIES IN TERMS OF 

(1) (2) 

Control Control 
Series 1. Boys Girls 

Roll hoops -5.27 -2.55 

Play jackstones -1.17 -4.04 

Spin tops -0.83 -0.83 

Play jackstraws -1 .1 7 -1 .1 7 

Walk on stilts -1.17 -1.85 

Fly kites -1.17 -2.73 

Play with bow and arrow -2.55 -2.02 

Coast or toboggan +0.67 0.00 

Ride bicycle +5.95 +0.67 

Skate -2.37 -2.91 

Ski -0.50 -0.67 

Hike +1.68 +2.20 

Do garden work -1.00 -3.84 

Dance 0.00 -2.20 

Shoot +3.84 -2.20 

Fish +1.17 -1.85 

Swim +1.85 -0.17 

Ride horseback +3.84 -0.33 

Row a boat +1.00 -0.83 

Hunt +3.84 -1.34 

Do plain t sewing +4.24 +0.33 

Cook a meal +1.85 +3.65 

Knit, crochet or do fancy work -0.50 +3.46 

Use tools +2.91 -1.17 

Work with machinery +3.84 -1.00 



TAN. < 


TAN. < 


6 


(3) 


(4) 


Gifted 


Gifted 


Boys 


Girls 


-1.85 


-2.73 


+0.83 


-1.68 


-2.20 


-1.51 


-2.55 


-2.55 


+1.51 


-1.17 


-2.20 


-2.02 


-3.28 


-1.00 


+1.34 


+0.33 


+1,00 


+2.37 


-0.17 


-4.85 


-0.50 


+0.17 


+2.91 


+1.68 


-1.51 


-1.68 


+1.00 


+1.34 


+1.17 


-0.17 


+0.83 


-0.17 


+4.44 


+2.37 


-1.17 


-0.83 


+1.68 


+1.17 


-0.67 


-0.17 


-1.00 


-3.65 


-0.33 


+4.04 


-0.83 


-2.20 


+2.02 


-1.00 


+1.00 


-0.17 



PLAY INTERESTS 



417 



TABLE 156 Continued 



Series 2. 

Play tag 

Ring around the rosy 

London Bridge 

Farmer in the dell 

In and out the window 

Hide and seek 

Hopscotch 

Drop the handkerchief 

Blindfold 

Postoffice 

Cat and mouse 

Red Rover 

Puss in the corner 

Pom-pom pull-away 

Blackman 

Marbles 

Duck on rock 

Follow the leader 

Anty over 

Dare base or prisoner's base 

Snap the whip 
Tug of war 
Roly-poly 
Jump the rope 
Leapfrog 

Fox and hounds 
Fox and geese 
Shinny 
Croquet 
Bowling 

Wrestling 

Baseball 

Racing or jumping 

Handball 

Soccer 

Boxing 

Tennis 

Volleyball 

Basketball 

Football 



(1) 


(2) 


(3) 


(4) 


Control 


Control 


Gifted 


Gifted 


Boys 


Girls 


Boys 


Girls 


-5.49 


-4.64 


-4.64 


-6.91 


-2.91 


-6.66 


-1.51 


-5.06 


-2.73 


-6.18 


-2.37 


-7.43 


-5.72 


-6.91 


-2.02 


-6.66 


-2.73 


-4.04 


-1.17 


-5.06 


-2.91 


-6.42 


-3.28 


-3.84 


-2.55 


-7.71 


-2.73 


-7.43 


-4.04 


-6.42 


-2.02 


-6.91 


-1.68 


-4.44 


-2.20 


-5.06 


-1.17 


-1.00 


-0.67 


-1.68 


-6.18 


-9.19 


-2.02 


-4.44 


-1.68 


-1.68 


-1.51 


-1.68 


-2.55 


-5.27 


-3.46 


-7.43 


-2.91 


-1.34 


+0.83 


-2.02 


-0.17 


-2.02 


-0.67 


-0.67 


-1.00 


-1.17 


-3.65 


-o.sa 


-1.00 


-0.83 


-1.00 


0.00 


-3.84 


-5.27 


-2.73 


-3.65 


-0.83 


-1.51 


-0.50 


+1.51 


-1.00 


+3.09 


-0.83 


+1.17 


-3.46 


-2.02 


-0.83 


-2.02 


-3.84 


-3.09 


-1.17 


-2.02 


-2.02 


-2.55 


-0.50 


-1.34 


-4.85 


-5.06 


-1.00 


-6.66 


-2.02 


-5.06 


-1.84 


-4.24 


-0.83 


-0.33 


-1.17 


-0.17 


-1.85 


-3.46 


-1.00 


-1.68 


+0.17 


-1.00 


-1.85 


-0.67 


+0.50 


0.00 


+0.67 


+1.85 


+0.17 


-1.34 


-0.33 


0.00 


+1.85 


-1.68 


-0.67 


-0.50 


+2.02 


+5.06 


+4.04 


+4,04 


+0.50 


-3.09 


+0.33 


-2.73 


-1.34 


-1.51 


+5.06 


+2.20 


+2.37 


-0.83 


+1.17 


+1.17 


+1.85 


-1.17 


0.00 


-0.17 


+1.68 


+1.68 


+2.20 


+3.65 


-2.37 


+3.84 


0.00 


+2.37 


+2.37 


+2.02 


+3.84 


+3.46 


+1.85 


-0.50 


+3.09 


-0.83 



418 



TRAITS OF GIFTED CHILDREN 



TABLE 156 Concluded 

Series 3. 

Play with dolls 

Play "dress up" 

Play house 

Play store 

Play school 

Simon says thumbs up 

Play church 

Dominoes 

Crokinole 

Parchesi 

Tiddledy-winks 

Backgammon 

Authors 

Snap 

Guessing games 

Cards 

History cards 

Geography cards 

Charades 

Anagrams or word building 

Solve puzzles 

Checkers 

Chess 

Pool 

Billiards 

As four maturity ratings have been computed for each 
activity (CB, CG, GB, GG), it was possible to secure the fol- 
lowing intercorrelations : 

Maturity Indices Based on 

Control Boys vs. Gifted Boys 

Control Girls vs. Gifted Girls 

Control Boys vs. Control Girls 

Control Boys vs. Gifted Girls 

Gifted Boys vs. Gifted Girls 

Gifted Boys vs. Control Girls 

These intercorrelations run much as one might expect, 
except for the lower correlation (.56) between control and 
gifted boys, as compared with the high correlation (.91) be- 
tween control and gifted girls. 

The norms of maturity indices must of course be based 
upon the control group. In order to have such norms in a 
form convenient for use, the maturity indices of Table 156 



(1) 


(2) 


(3) 


(4) 


Control 


Control 


Gifted 


Gifted 


Boys 


Girls 


Boys 


Girls 


-1.51 


-10.57 


0.00 


-11.34 


-1.34 


-7.98 


-0.67 


-9.85 


-3.84 


-9.51 


-0.67 


-11.34 


-3.84 


-6.91 


-1.68 


-8.87 


-3.84 


-6.91 


-2.02 


-9.19 


-2.20 


-4.04 


-3.46 


-5.27 


-0.83 


-5.27 


-0.33 


-1.68 


-1.68 


-1.34 


-1.85 


-2.02 


-0.83 


-0.67 


+1.51 


-0.33 


-1.68 


-0.17 


-1.17 


+1.17 


-3.65 


-2.20 


-2.37 


-1.85 


-0.67 


-0.67 


0.00 


0.00 


-1.34 


-0.17 


-0.50 


+0.17 


-2.55 


-1.85 


-1.17 


-0.17 


-0.83 


-1.34 


-2.37 


-0.50 


-1.85 


+0.83 


-1.00 


-0.33 


0.00 


-1.17 


-1.85 


-1.85 


-0.67 


-1.00 


-0.83 


-0.33 


-0.67 


-0.67 


-0.67 


+2.73 


-1.17 


-0.83 


-1.51 


-0.33 


+2.02 


+1.34 


+0.50 


+1.17 


+1.34 


+0.17 


+0.50 


-0.17 


-0.67 


-0.33 


+2.37 


-0.33 


-0.83 


-1.00 


+0.50 


-0.33 


0.00 


-1.00 


-0.67 


-0.17 



Pearson r 

.56 .048 

.91 .012 

.58 .048 

.59 .047 

,57 .048 

.45 .057 



PLAY INTERESTS 

TABLE 157 
MATURITY INDICES OF ACTIVITIES, CONTROL GROUP 



419 



Boys 


Girls 


Boys Girls 


Boys Girls 


7 


13 


Roll hoops 


7 


10 


Play tag 


13 


1 


Play with dolls 


13 


11 


Play jackstones 


11 


7 


Ring around rosy 


13 


5 


Play "dress up" 


14 


16 


Spin tops 


11 


7 


London Bridge 


9 


3 


Play house 


13 


15 


Play jackstraws 


6 


6 


Farmer in the 


9 


6 


Play store 


13 


14 


Walk on stilts 






dell 


8 


6 


Play school 








11 


11 


In and out 


















window 








13 


13 


Fly kites 


11 


7 


Hide and seek 


12 


11 


Simon says 


11 


14 


Play with bow 


11 


5 


Hopscotch 






thumbs up 






and arrow 


9 


7 


Drop 


14 


9 


Play church 


16 


17 


Coast or 






handkerchief 


13 


15 


Dominoes 






toboggan 


13 


10 


Blindfold 


14 


16 


Crokinole 


25 


18 


Ride bicycle 


13 


16 


Postoffice 


13 


17 


Parchesi 


12 


13 


Skate 














15 


16 


Ski 


6 


5 


Cat and mouse 


10 


14 


Tiddledy-winks 


18 


21 


Hike 


13 


15 


Red Rover 


14 


16 


Backgammon 


14 


11 


Do garden work 


11 


9 


Puss in corner 


13 


17 


Authors 


15 


14 


Dance 


11 


15 


Pom-pom pull- 


11 


14 


Snap 


21 


14 


Shoot 






away 


14 


15 


Guessing games 








15 


15 


Blackman 








17 


14 


Fish 


14 


15 


Marbles 


12 


19 


Cards 


18 


17 


Swim 


14 


16 


Duck on rock 


15 


15 


History cards 


21 


17 


Ride horseback 


9 


9 


Follow leader 


14 


16 


Geography cards 


17 


16 


Row a boat 


14 


15 


Anty over 


14 


16 


Charades 


21 


15 


Hunt 


14 


22 


Dare base 


13 


16 


Word building 


22 
18 


18 
23 


Do plain sewing 
Cook a meal 


10 
9 


21 
12 


Snap the whip 
Tug of war 


19 
17 


19 
18 


Solve puzzles 
Checkers 


15 


23 


Knit or crochet 


12 


13 


Roly-poly 


14 


17 


Chess 


20 


15 


Use tools 


8 


9 


Jump the rope 


14 


16 


'Pool 


21 


16 


Work with 


12 


9 


Leapfrog 


15 


16 


Billiards 






machinery 




















14 


17 


Fox and hounds 














12 


12 


Fox and geese 














16 


16 


Shinny 














16 


17 


Croquet 














16 


15 


Bowling 














18 


15 


Wrestling 














19 


25 


Baseball 














16 


12 


Racing and 


















jumping 














13 


15 


Handball 














19 


16 


Soccer 














18 


15 


Boxing 














18 


20 


Tennis 














12 


23 


Volleyball 














19 


21 


Basketball 






- 








18 


17 


Football 









420 TRAITS OF GIFTED CHILDREN 

were transmuted into equivalent numerical indices ranging 

from 1 to 25, by the use of values, as was done in deriving 

the masculinity indices. These are given in Table 157. In 
this table, 13 means that the general trend of the curve is 
toward neither greater nor less preference with increasing 
age. Indices above 13 mean that preference increases with 
age; indices below 13, that it decreases with age. 

MATURITY RATINGS OF CHILDREN 

The main purpose of the maturity indices of the activities 
was to serve as a basis for deriving maturity ratings of the 
individual children with respect to their play interests. The 
method of securing these ratings was exactly the same as 
that used in securing masculinity ratings of the children. 
The distributions, means, and S.D's of the maturity ratings 
of the children are shown in Tables 158 and 159. 

The extent to which the means in Tables 158 and 159 
show significant differences in the maturity ratings of the 
control and gifted groups, may be seen in Table 160. 

The differences may be regarded as significant for the 
boys at all ages except 13, and for the girls at all ages except 
8 and 9. 

Figure 26 shows the mean maturity ratings for the gifted 
and control groups, separately by sex. It will be noted that 
both gifted.boys and gifted girls are considerably more ma- 
ture in their play interests than are the control groups of 
corresponding age. 

Table 161 shows the per cent of gifted children in vari* 
ous percentile ranges of the control group of corresponding 
sex. It will be noted that a much larger proportion of the 
gifted than of the control children make high maturity rat- 
ings, especially the gifted boys of 6 to 10 years. 

SOCIABILITY AND ACTIVITY RATINGS OF THE CHILDREN 

It will be recalled that the activities of column 1 are in 
the main non-social; those of column 2, social and in most 
cases competitive; those of column 3, mildly social and 
"quiet." It will be of interest, therefore, to note what pro- 
portion of a given child's expressed preferences are in col- 
umn 2. A high proportion would seem to indicate greater 



PLAY INTERESTS 



421 



rHOOCDCO^OOOOOCOCsltH 



CO-^tJ 



rHOQO CftCOOOTH 



I 



o 

a 



)H 
O 







I 

I 

i 



CO -^f CO N rH rH 



a 



CO LO XO CO O TH TH 



COlO 

TH C'QI 



c^ us t oo c<i o TH 



s? 

C3COO 



cqo5 
O373O 



TH "^ t^- CO CM 



TH ^3 O3 00 t" CO iO ''J* CO C*3 i^ ^S 
(NSMi-HiHl-dHrH^Hi-li-li-l^-l 



422 



TRAITS OF GIFTED CHILDREN 



33 



THi iTHT^COCOOOf-OOrH 



I 



<M IM t- CO 



o 



toco c4 

CM rH 



PH 



H 



gs 



Tt< rH 

Tt^rH 



O' 



CO Th tC C<1 O3 CO rH 



! 



PLAY INTERESTS 



423 



TABLE 160 
SIGNIFICANCE OF MATURITY DIFFERENCES IN TABLES 158 AND 159 



Control and Gifted Boys 


Control and Gifted Girls 




Diff. 





Diff. 


Diff. 





Diff. 




between 


of the 


divided 


between 


of the 


divided 


Age 


means 


diff. 


by a 


means 


diff. 


by a 


8 


2.21 


0.361 


6.12 


0.49 


0.523 


0.94 


9 


1.21 


0.287 


4.22 


0.57 


0.420 


1.36 


10 


1.49 


0.275 


5.42 


1.04 


0.508 


2.05 


11 


0.66 


0.257 


2.57 


1.45 


0.413 


3.51 


12 


0.86 


0.380 


2.26 


1.80 


0.484 


3.72 


13 


0.43 


0.382 


1.13 


1.06 


0.455 


2.33 



Total 



0.65 



0.149 



4.29 



0.63 



0.245 



2.57 



FIGURE 26 
MEAN MATURITY RATINGS BY AGE, GIFTED AND CONTROL GROUPS 




11 



6-7 8 



10 11 12 15 14 15-17 



424 



TRAITS OF GIFTED CHILDREN 



I 



oo oo o 



CO 
CO 



en oo TH 

CM if* TH 



8 

I 



OO TH CM CD 

lO CO 



O 

3 ! 

(g 

1 " 

H W 



CM ^ IA CO 
SO Oi TH CM 



OO O CM CD O 

c^ r^ TH co CD 



I * 

P-< 

CO 5 



CM 00 
CM CD 



OO 
CM 



00 
CM 



TH CO 

OO t"^* 



<! 3 



52 5 | 
S fc 3 



O CD CO CD 



H 

s 



o 



TH CD 00 OO 



00 
00 



^ CM 



T)h t*^ OO CO CD 

OS ITS OO CO O 
TH O 



CO 

O 

5 

i 



i 

o 



CD CO OO TH OO CD 

CD ^ CO I> ^ CD 

TH CM irt 



tf 
P 







O 

S 



o *- o o 

#3 g g ^ ^J ^ ta 

5rd *- ^3 *- 2^ 

2S" W ig> g 2 g I 

jx, o JH J2^ J2I H- 3 



2 

a 



PLAY INTERESTS 425 

sociability; a low proportion, less sociability. Accordingly, 
for each child the three columns were scored separately, us- 
ing the method of weighting by one cross, two crosses, and 
three crosses, described on pages 394-395; then the ratio of 
score on column 2 to total score on the three columns was 
calculated and used as a sociability rating of the child. In 
the same way, the proportion of a child's expressed prefer- 
ences in column 3 ("quiet" plays) may be used as an activity 
rating. In this case, the greater the per cent of score on 
column 3, the less the activity interest. 

In deriving both sociability and activity ratings, Exercise 
1 was ignored and the scores on Exercises 2 and 3 were com- 
bined. 

Table 162 gives the classification of sociability ratings 
assigned to the gifted for each age and sex group. It will 
be noted that from a third to a half of the sociability ratings 
of the gifted fall below the lower quartile of the control. 
The difference is large enough to be significant. 

Table 163 gives the corresponding data for the activity 
ratings of the gifted. Here again the gifted show a lower 
average rating than the control group, indicating that they 
are more interested in games of the "quiet" type. 

TESTIMONY REGARDING EXPERIENCE AND SKILLS (EXERCISE 4) 

Exercise 4 requires the child to answer 45 questions re- 
garding his experience and skills. These questions, if an- 
swered truthfully, would throw considerable light on the 
degree to which gifted children differ from average children 
with respect to their experience and interest in wholesome 
activities. Inasmuch as gross overstatement is possible, and 
in a certain proportion of cases very probable, the scores on 
this exercise cannot be relied upon for a rating of individual 
children. They can be used, however, for the comparison of 
groups, provided the relative tendency of the various groups 
to overstate is known. This information is available from 
the Overstatement Test described in Chapter XVII. It is 
there shown that gifted children of all ages are less inclined 
than the control group to overstate their knowledge. There is 
probably little risk in assuming that this tendency holds also 
for the test with which we are here concerned, and that in 
consequence any differences found between the gifted and 



426 



TRAITS OF GIFTED CHILDREN 







H 

g 


cated Percentile Banges 


tM 

TH 


'T CO "3 

--3 _J < 

o S 5 

I > 

* S S 

g oq <?* 

o m 5J 

"* v 

S -i 

S 2 S 


C7 

CO 

co 

CO 

CO 



rH 

CO 
CM 


t* *; : 

& S 3 eg 

^ lO CO CM 
rj cf^ lO rH 
O CM CO ^ 
r-t 

g o^ r^ co T-H 

a p* S S 5 

W 3 

5 'S ^ ^ "^ 

^ 35 CM CM -*i 
M *^ ^ CM CM S 




^ 












e i s 


CM 

CO 

T-t 




03 


.2 




S. "^ ^ s * 
o * ^ 
CQ oo "> 


GO 

CO 


co >, ^- ^^ ^** 

50 fe 'S 1" ^ oq co 
"- 1 c3 c* CM ^ 


S 

3 

H 


RATING 


K 

i 




3 S 


CO 


M t ^ * ^ 
3 O t*> 

g "^ -g ^ ^ ^ 

r^ 13 ^ Cft xo U3 




e 


-<j 


Oi 






tx * " 




IH 
3 


^ 




? ^- "fe*- 

* co o 


r- 


HO SL ^N^ "fe^ ^* 

> <M O "^ <^ ^ 




S 


S 




OQ 


CO 


H ^, rH CO UO 







S 








-< S ooo 




OO 






W ^^ ^^ 


Q^- 


,fS ^* "o^ ^^ 











^5 10 ^f 


rH 


o3 .^ ^ *<; ^ 










O CM* ^" 

rH CO 


S 


0* C? CO O CO 









00 






00 






I 




^ "^ 
CQ rH CO 


o 


g 1 CO CM 

qa co <=> co 










rH CO 


\a 


CM VO CM 








^ 


2 oo o 


CM 


c* .tj oo o^ co 








TJ 


O * o 


CC5 


rt O ^5 TH co 








j^ 


TH 




J5 CO CO 








05 






C8 








CO 






s c, o o 










en ^s. ^v. 














o* 9 ? 


^ 


^ t^ oq co 








en 
co 


ft "* 


^ 


" 5 s S 








bO 

PJ 
rt 


3 

O -vft 

t* ** "^. 





taa pj 
g o ^ Q ^ 










1 jj? g 


CM 


3 S CM 








"S 

g 


rf3 3 
U t* ^O 


1 


ll 1 i P 











**H >r5 ^SJ 
W ^ 


2 


S 's S 3 



PLAY INTERESTS 



427 



control are more unfavorable to the gifted than they ought 
to be. Table 164 gives the means and S.D's by age, sex, and 
intelligence, on the basis of 1 point for each question an- 
swered by underlining "yes." 

TABLE 164 
CLAIMS REGARDING EXPERIENCE AND SKILLS 





6-7 


8 


9 


10 


11 


12 


13 


14 


15-17 




G 


C G 


G 


C G 


C G 


C G 


C G 


C 


C 


Boys Mean 


6.8 


11.0 12.0 


15.8 12.8 


15.1 15.6 


17.6 15.3 


15.6 16.8 


19.9 20.2 


19.0 


18.2 


S. D. 


4.8 


4.5 5.9 


8.2 6.9 


7.3 6.9 


9.0 6.1 


6.5 6.5 


5.6 4.8 


5.9 


5.1 


Girls Mean 


8.1 


9.6 6.1 


8.9 8.7 


11.8 9.9 


11.2 10.2 


12.8 12.9 


10.8 10.9 


12.0 


11.0 


S. D. 


5.4 


6.6 S.8 


6.2 5.6 


6.2 4.8 


6.6 4.2 


6.1 5.1 


4.6 3.7 


5.1 


4.5 



From the above data it would seem that gifted children 
have had, on the whole, about as rich experience along the 
lines to which these questions relate as the control children 
have had, possibly richer, if we allow for the greater tend- 
ency of control children to overstate in matters of this kind. 
However, interesting differences were found on individual 
items when the per cents of affirmative replies were cal- 
culated for the control and gifted groups. The largest of 
these differences are as follows : 



Question 

Did you ever shoot any game? 

Have you ever set up electrical 
apparatus? 

Can you read the time from a sun dial? 

Have you ever hiked eight miles 
in a day? 

Have you ever milked a cow? 
Have you ever hitched up a horse ? 

Have you ever been elected to any 
office or special honor? 

Have you ever been captain of an 
athletic team? 



PER CENT OF AFFIRMATIVE REPLIES 

Control Gifted Control Gifted 

Girls 



Boys 
54 

19 

28 

51 
57 
52 

21 
24 



Boys 

31 

39 
58 

69 
27 
27 

53 
34 



Girls 

17 

15 

34 
29 
28 

18 
15 



42 

47 
17 
10 

53 
21 



428 TRAITS OF GIFTED CHILDREN 

EXERCISES 5, 6, AND 7 (PLAY INFORMATION) 

Exercises 5, 6, and 7 (see page 390 ff .) are tests of play in- 
formation and relate chiefly, though not entirely, to the 90 
activities of Exercises 1, 2, and 3. The score in each exercise 
is the number of correct responses minus half the number of 
wrong responses; that is, score /? %W. This formula 
allows for the factor of chance success from guessing. In 
order to increase the reliability of results, the scores on the 
three exercises were combined into a total score. Means and 
S.D's of the control and gifted groups by age and sex are 
given in Table 165. 

TABLE 165 

MEANS AND S.D's FOR PLAY INFORMATION BY AGE, SEX, 
AND INTELLIGENCE 





6-7 

G 


8 

C G 


9 

O G 


10 
G 


11 

G 


Boys Mean 
S.D, 
Girls Mean 
S.D. 


15.3 
11.2 
15.6 

8.5 


14.8 32.5 
5.8 15.0 
8.0 23.6 
4.4 11.4 


17.8 49.8 
11.7 15.5 
13.3 33.1 
9.5 14.6 


22.7 58.0 
15.6 15.3 
18.3 44.3 
9.1 15.4 


35.6 70.1 
21.3 13.5 
31.3 53.5 
15.4 15.1 




12 

G 


13 

G 


14 
o 


15-17 






Boys Mean 
S.D. 
Girls Mean 
S.D. 


46.1 77.3 
19.3 11.7 
33.8 62.6 
16.4 12.5 


50.0 83.7 
26.9 8.0 
49.1 71.7 
18.7 12.0 


50.7 
23.7 
42.9 
19.8 


32.4 
24.8 
38.3 
14.3 



The age means for the control boys were smoothed and 
used as norms for calculating the information quotients of 
the individual gifted boys, just as intelligence quotients are 
calculated. As the sex differences are very large, a separate 
norm was derived for the girls. Table 166 shows the dis- 
tribution of play information quotients for the gifted chil- 
dren. 

The mean quotient of 136.8 shows that gifted children, 
age for age, possess enormously more information about 
plays, games, and amusements than do unselected children. 
In fact, their superiority in play information is almost as 
marked as their superiority in the Stanford Achievement 



PLAY INTERESTS 



429 



Tests. It is of course unsafe to draw any inferences regard- 
ing play habits from data on play knowledge, but it is inter- 
esting to know that whatever the play habits of the gifted 
may be, such children are at any rate not lacking in play in- 
formation. The results here lend greater credence to the 
data from Exercises 1, 2, 3, and 4. 

TABLE 166 
PLAY INFORMATION QUOTIENTS OF GIFTED CHILDREN 



Gifted Boys 


Gifted Girls 


Total, Gifted 


Quotient 


N 


Per Cent 


N 


Per Cent 


N 


Per Cent 


190-200 






1 


0.4 % 


1 


0.2 % 


180-190 






4 


1.6 % 


4 


0.7 % 


170-180 


12 


3.9 % 


4 


1.6 % 


16 


2.9 % 


160-170 


32 


10.5 % 


13 


5.2 % 


45 


8,1 % 


150-160 


51 


16.7 % 


31 


12.3 % 


82 


14.8 % 


140-150 


62 


20.3 % 


39 


15.5 % 


101 


18.2 % 


130-140 


51 


16.7 % 


60 


23.9 % 


111 


19.9 % 


120-130 


51 


16.7 % 


46 


18.3 % 


97 


17.5 % 


110-120 


29 


9.5 % 


29 


11.5 % 


58 


10,3 % 


100-110 


11 


3.6 % 


14 


5.8 % 


25 


4.5 % 


90-100 


3 


1.0 % 


9 


3.6 '% 


12 


2.2 % 


80- 90 


1 


0.3 % 


1 


0.4 % 


2 


0.4 % 


70- 80 


2 


0.7 % 






2 


0.4 % 


Total 


305 




251 


100 % 


556 




Mean 




138.99% 




134.3 % 




136.87% 


S.D. 




18.15% 




19.32% 




19.24% 



It was originally intended that the play information 
scores should also be used as a check against the child's hon- 
esty of report in Exercises 1, 2, and 3. This task was aban- 
doned, partly because it was too laborious, but chiefly for 
the reason that the relatively small number of activities 
marked by many children in Exercises 1, 2, and 3 make the 
check very unreliable. 

HOME AND SCHOOL DATA ON PLAY LIFE 

In addition to the data secured by the use of the ques- 
tionnaire-test, considerable supplementary information on 
play life was furnished by the Home and School Information 
Blanks. In each certain questions were asked regarding the 
amount of time spent in play, age and sex of companions 
preferred, attitude toward and popularity with other chil- 



430 TRAITS OF GIFTED CHILDREN 

dren, etc. In most cases a graded rating was called for, on 
the same general plan, in both home and school reports. In 
the case of the gifted children it is therefore usually possi- 
ble to check one report against the other. 

Home and School Blanks were not filled out for the con- 
trol group which was given the questionnaire-test on plays, 
games, and amusements. However, as stated on page 177 ff., 
School Information Blanks were filled out by teachers for a 
control group of approximately 600 children in Los Angeles, 
San Francisco, and Oakland. (Control Group A.) It will be 
recalled that these children were selected to represent as 
nearly as possible the average school child in the cities, 
schools, and neighborhoods from which the gifted children 
came. 

Since this group of control children does not go below 
age 8, wherever age affects the results the comparison is 
somewhat unfair to the gifted. The reports were all tabu- 
lated by age, but as there was little evidence of age influence 
on most of the questions reported upon, the ages have or- 
dinarily been combined in the summaries given in the fol- 
lowing pages. The proportion in each group for whom re- 
ports were received will be stated in each case. Unless other- 
wise stated, the tabulated results are in terms of per cents 
for whom the given question was answered. 

Does child play with other children very much, average 
amount, little? Underline. School Blank: IV, 10. Answered 
for 91 per cent of CB, 93 per cent of GB, 92 per cent of CG, 
and 89 per cent of GG. 

Very Average 

much amount Little 

Control Boys 33% 55% 12% 

Gifted Boys 20% 64% 16% 

Control Girls 33% 57% 10% 

Gifted Girls 24% 64% 12% 

These figures indicate that the gifted child plays alone 
somewhat more than do normal children similarly situated. 
The difference, however, is small. 

Average hours a week spent with other children during 
the last year (out of school). Home Blank: II, 29. Answered 
for 86 per cent of boys and 89 per cent of girls. 



PLAY INTERESTS 431 

Average Hours a Week 

35 or 28 21 14 7 

more to 35 to 28 to 21 to 14 to 7 

Gifted Boys 9% 12% 23% 20% 28% 8% 

Gifted Girls 8% 7% 16% 25% 34% 10% 

Comparable data are not available for the control group, 
but the average of about two and three-quarter hours a day 
for gifted boys and two and a quarter hours a day for gifted 
girls, spent in play with other children out of school, would 
not seem to be a bad record. 

Prefers playmates who are much older, older, same age, 
younger, much younger. School Blank: IV, 11. Home Blank: 
II, 30. Home reports for 91 per cent of gifted; school reports 
for 87 per cent of gifted and 90 per cent of control. As no 
significant sex differences were found, the sexes were com- 
bined. 

Much Same Much 

older Older age Younger younger 

School Blank: Control 1.2% 7.9% 86.5% 3.8% 0.6% 

" Gifted 4.3% 20.8% 70.8% 3.3% 0.8% 

Home Blank: Gifted 4.2% 30.4% 61.2% 4.0% 0.2% 

The school reports a much larger percentage of gifted than 
control children who prefer older playmates, and the home 
reports for the gifted agree fairly well with those from the 
school. This is probably due in part to the fact that the 
gifted child is usually associated in school with children a 
year or two older than himself, and in part to a tendency 
for mental ages to seek their level. 

Prefers playmates of the same sex, opposite sex, or no 
preference. Underline. School Blank: IV, 12. Home Blank: 
II: 32. Home reports for 98 per cent of gifted; school re- 
ports for 89 per cent of gifted and 91 per cent of control. 



20% 
33% 
43% 

The significant difference Here between the control and 
gifted is the much less marked tendency of the gifted to 







Same 


Boys 
Opposite 


No Pref. 


Same 


Girls 
Opposi 


Control 


: school 














report 


79% 


4% 


17% 


75% 


5% 


Gifted: 


school 














report 


76% 


1% 


23% 


65% 


2% 


Gifted: 


home 














report 


67% 


3% 


30% 


54% 


3% 



432 TRAITS OF GIFTED CHILDREN 

make sex distinctions. This is especially noticeable in the 
case of the gifted girls, and is in harmony with the fact that 
the masculinity ratings based on play interests were higher 
for gifted than for control girls. Home and school agree in 
reporting the girls (both gifted and control) considerably 
more tardy than boys in developing a preference with re- 
gard to sex of playmates. This is shown in the following per 
cents reported as having "no preference" at various ages : 

No PREFERENCE FOR SEX OF PLAYMATE 

Age Age Age Age Age 

2-7 8-10 10-12 12-14 14-15 

School Blank: Control Boys 29% 11% 11% 14% 

Girls 20% 22% 20% 18% 

" Gifted Boys 30% 23% 20% 24% 

Girls 35% 25% 29% 34% 

Home Blank: Gifted Boys 60% 29% 21% 19% 

Girls 35% 37% 42% 54% 

Reading the above figures from left to right it is seen that for 
girls the absence of any sex preference is more common at 
12-14 than at 8-10, and that the reverse holds for the boys. 

Is child's companionship especially sought, rather 
avoided, neither? Underline. School Blank: IV, 13. Answers 
were received for 91 per cent of both groups. 

Boys Girls 

Especially Rather Especially Rather 

sought avoided Neither sought avoided Neither 

Control 32% 3% 65% 38% 3% 59% 

Gifted 29% 5% 66% 37% 4% 59% 

The figures for control and gifted agree fairly closely; more 
girls than boys in each group are said to be especially sought 
for companionship. This finding is supported to some extent 
by the trait ratings presented in Chapter XVIII. The follow- 
ing questions also throw some light on popularity and social 
adaptability. The following figures for three different age 
levels show that with both gifted and control the per cent of 
children "rather avoided'* decreases after 11 years: 

Ages 8 and 9 Ages 10 and 11 Ages 12 and 13 

Control Boys 5% 4% 

Control Girls 1% 4% 

Gifted Boys 7% 7% 2% 

Gifted Girls 1% 5% 4% 



PLAY INTERESTS 433 

When can't have his way, cries or gets angry often, occa- 
sionally, rarely. Underline. School Blank: IV, 14. Answered 
for 81 per cent of gifted and 82 per cent of control. 

Boys Girls 

Often Occasionally Rarely Often Occasionally Rarely 

Control 5% 21% 74% 6% 12% 82% 

Gifted 7% 14% 79% 4% 14% 82% 

The differences here are too small to have any significance. 

Is teased by others very frequently, frequently, occa- 
sionally, rarely, never. Underline. School Blank: IV, 15. 
Home Blank : II, 34. School reports for 88 per cent of both 
groups; home reports for 95 per cent of gifted. 

Very 
Boys Frequently Frequently Occasionally Rarely Never 

School Blank: CB 0% 6% 19% 46% 29% 
School Blank: GB 3% 4% 21% 30% 42% 
Home Blank: GB 3% 8% 34% 38% 17% 

Girls 

School Blank: CG 1% 7% 18% 34% 40% 

School Blank: GG 1% 4% 12% 36% 47% 

Home Blank: GG 3% 8% 31% 37% 21% 

The above figures do not show any consistent trends, ex- 
cept that teachers more often than parents report children 
as never teased. There is no evidence from these reports that 
gifted children tend more often than others to be socially 
maladjusted. The proportion "rarely" or "never" teased re- 
mained in all groups practically constant for all ages, the 
gifted girls running higher than control girls, and the gifted 
boys somewhat lower than the control boys. 

Is child considered by others as "queer" or different? If 
so, in what way? Home Blank: II, 35. School Blank: IV, 
16. Home reports were received for 89 per cent of gifted 
boys and for 92 per tent of gifted girls; school reports for 84 
per cent of control boys, and 87 per cent of control girls, 89 
per cent of gifted boys, and 83 per cent of gifted girls. 

Boys Girls 

No Yes No Yes 

School Blank: Control 95% 5% 95% 5% 

" Gifted 88% 12% 93% 7% 

Home Blank: Gifted 90% 10% 92% 8% 



434 TRAITS OF GIFTED CHILDREN 

This would indicate that gifted children somewhat more 
often than control children are considered "queer" or differ- 
ent, although the absolute number is not large. There is 
close agreement of the reports from home and school on the 
gifted group. The question, m what way? brought the fol- 
lowing explanations of the adjustment difficulties found in 
the gifted group : 



Home Blank 
Boys Girls 



School Blank 
Boys Girls 



Number of cases reported 

No explanation given 

"Considered remarkable," "brighter," 

"knows more" - 

Acts more mature, "old acting" 
Serious conduct 
Would rather read than play 
Talkative 
Domineering 
Independent 
Conceited, "aloof" 
Lacks physical courage 
Quiet, too refined, (boys call him "sissy") 
Doesn't care for outdoor games 
Not rough acting 
Physical disabilities prevent participation 

in games 
"So sensitive" 
Dramatizes everything 
Original hates sameness 
Japanese 

Dislikes other children no chums 
Quarrelsome and antagonistic 
Babyish (plays in school) 
Fat and clumsy 
"Doesn't fit in" 



20 
2 

6 
4 
1 
1 
1 
1 
1 
3 



30 
3 

10 



16 
2 

7 
1 

2 
1 



1 
1 
1 



33 
1 

8 
1 

4 
1 
3 



2 
2 
1 
1 
1 
1 



One cannot assume that the reasons given are always 
the true ones, although the agreement between the home and 
school reports suggests that general tendencies may be re- 
vealed. If so, it is probable that in a majority of cases the 
difficulty is not serious enough to endanger greatly the child's 
future. Less than half the explanations mention traits that 
are seriously unfavorable. 

Prefers to play indoors, outdoors, or no preference. Un- 
derline, Home Blank: II, 33. Answered for 97 per cent. 



PLAY INTERESTS 435 

Boys Girls 

Outdoors Indoors No Pref. Outdoors Indoors No Pref. 

Gifted 63% 5% 32% 53% 8% 39% 

We have no control data on this question, but the results are 
about what one might expect from children in general; 5 
per cent of boys and 8 per cent of girls prefer indoor play. 
No age differences were noted. 

Are play interests normal? If not, explain. School Blank: 
IV, 17. Answered for 93 per cent of gifted and 95 per cent of 
the control. 

Boys Girls 

Yes No Yes No 

School Blank : Control 95% 5% 95% 5% 

School Blank: Gifted 90% 10% 97% 3% 

Gifted boys are slightly more often abnormal in their 
play interests than control boys, according to these figures. 
Explanation of these "abnormalities" was given for the 
gifted children as follows : 

Girls Boys 

Number of cases reported 9 32 

No explanation given - 2 

"Prefers reading to play" 2 14 

"Does not want to play" 4 9 

"Not interested in games" 1 5 

"Reserved," "bashful" 1 1 

"Prefers one or two companions" 1 

"Plays constantly with one little girl" - 1 

Has child had imaginary playmates? . . . Imaginary 
countries? Home Blank: II, 31. Imaginary playmates were 
reported as follows : 

Have Imaginary Playmates 
Girls Boys 

Number of cases reported 85 51 

No information given 24 8 

Definite personality 27 13 

Rather indefinite, but probable 3 1 

Fairies 4 1 

Imaginary animals, birds, etc. 1 3 

"Air friends" - 1 

"Makes up stories, incl. imaginary playmates" 1 - 

"Not often," "a few," "for short time" 4 1 

Imaginary opponent in games 1 6 

Assumes roles of characters in books 12 11 

Conversation with imaginary persons 8 7 

"Idealizes play animals and pillows" - 1 



436 TRAITS OF GIFTED CHILDREN 

A good many of these reports seem to be due to a mis- 
interpretation of the question, but it is probable that a fairly 
large proportion of the gifted children have had imaginary 
playmates. The information on these is rather meagre, but 
includes such statements as the following: 

"At the age of two, 'Don' and 'Gaul*. Very vivid." 
"From the age of twenty months to three years he played 
with imaginary animals. He loved them dearly and was 
never afraid." 

"At the age of two and three, 'Dibby/ who was blamed for 
his naughtiness." 

"A weaker boy who could be ordered around." 
" 'Oliver' and 'Stuart/ who played, ate, and slept with him. 
Very real." 

"Mr. and Mrs. Moon, Baby Moon, and their cook. Mrs. Card- 
board. She talks about them continually, and likes to take 
Baby Moon along when we go out." (A 3-year-old girl.) 
"Between the ages of three and four, 'Evelyn,' who did all 
the naughty things, but had to be petted.'* 
"Between the ages of three and five, Tin Can family of rela- 
tives. They came to see us, and she visited them." 
"Several, each named and never confused." 
"When three to five years old she used to say, 'Don't sit in 
that chair. Iris sits there/ " 

In the instances in which dates were given the imaginary 
playmates seem to have flourished when the children were 
from 2 to 5 years of age, many of them lasting through the 
whole period. There is some indication, too, that they occur 
more frequently in cases where the child has no real play- 
mates. 

The list of imaginary countries is summarized as follows : 

Have Imaginary Countries 

Girls Boys 

Number of cases reported 25 23 

"Wonderful places he has visited" - 1 

"Kyienien" Hills - 1 

Travel in general 1 5 

Fairy countries 3 2 

Oz Book type - 2 

Foreign countries (travel) 2 1 
"The Mountains" 1 



PLAY INTERESTS 437 

Have Imaginary Countries 
Girls Boys 

"Sailing in a Norwegian ship" - 1 

"Imaginary playmate, located variously" - 1 

"All sorts of places" 2 2 

"Fond of mythology" 1 

"The last book he has read" - 1 

"Where you can do everything you want, 

be successful in everything, see nice 

people, beautiful rivers, etc." 1 - 

No information given 14 4 

SUMMARY 

1. Preference ratings were assigned to 90 plays, games, 
and amusements, based on the extent to which gifted and 
unselected children say they like and practice them. Four 
such ratings, based, respectively, on control boys, control 
girls, gifted boys, and gifted girls, were derived. The inter- 
correlations of the ratings for these four groups were com- 
puted and found very high in each intelligence group (above 
.80) for like-sex groups, and very low (.18 to .35) for unlike- 
sex groups. 

2. The gifted children show measurably greater interest 
than the control in activities that require thinking and that 
are mildly social and quiet. They show slightly less prefer- 
ence than do the control group for competitive games. 

3. The sex differences in preferences expressed by the 
control group were made the basis of "masculinity indices" 
of the activities. This made possible the calculation of a 
"masculinity rating" for each child, based on the total mas- 
culinity values of the activities for which his preferences 
were expressed. 

4. The mean masculinity ratings of the gifted boys were 
slightly higher than those of control boys at all ages except 
13. The means of the gifted girls did not differ consistently 
from those of control girls. 

5. "Maturity indices" of the 90 activities were computed, 
based on the angle of their respective age curves of prefer- 
ence ratings. Four such indices were derived for each ac- 
tivity, based, respectively, upon control boys, control girls, 
gifted boys, and gifted girls. Intercorrelations of these ma- 



438 TRAITS OF GIFTED CHILDREN 

turity indices for the four groups ranged from .45 to .59, ex- 
cept for control girls vs. gifted girls, for which r.=.91. 

6. "Maturity ratings" of the children were then derived, 
based on the maturity indices possessed by the activities for 
which they had expressed preference. These ratings are con- 
sistently higher for gifted than for control children. This 
means that, in general, gifted children tend to prefer activi- 
ties which, among control children, show increasing rather 
than decreasing popularity for the ages 8 to 14. 

7. A "sociability rating" was derived for each child, based 
on the proportion of preferences which fell to social and 
competitive games. This tended to run somewhat lower for 
gifted than for control children. 

8. In the same way an "activity rating" was derived for 
each child, based on the proportion of preferences which 
fell to the quiet activities. These ratings were lower for girls 
than for boys, and also lower for the gifted than for the con- 
trol group. 

9. Testimony regarding 45 items of "experience" shows 
little difference between gifted and control in amount of 
such experience, but considerable difference in kind. The 
gifted have had much more experience along lines that in- 
volve intellectual activity. 

10. A play information test of 143 items was given to 
gifted and control groups. Norms were established and play 
information quotients were computed for the sexes sepa- 
rately. The mean play information quotient of the gifted was 
136. Girls scored somewhat lower than boys on this test. 

11. Additional data on play life were secured both from 
home and school for the gifted, and from the school for a 
control group. 

12. The gifted children play alone slightly more than do 
the control. However, the average gifted boy spends about 
two and three-quarter hours a day with other children (out 
of school), and the gifted girl about two and a quarter hours. 

13. Gifted children, oftener than control, prefer play- 
mates who are older than themselves. 

14. The gifted show much less sex preference than do 
control children in choice of playmates, and girls show far 
less sex preference than do boys. 



PLAY INTERESTS 439 

15. The gifted child's companionship is sought in school 
to about the same extent as that of the control child, although 
the gifted child is usually considerably younger than his 
classmates. 

16. There is little difference in the extent to which gifted 
and control children are teased by others or cry when they 
cannot have their way. However, somewhat more gifted 
than control are said to be regarded by other children as 
"queer" or "different." 

17. The play interests of gifted boys are somewhat more 
often said not to be normal than is the case with control 
boys. The reverse, however, holds for the girls. 

18. A good many gifted children have had imaginary 
playmates or imaginary countries, but comparative data on 
this point are not available for unselected children. 



CHAPTER XV 



READING INTERESTS 1 

The data of this chapter are based upon (1) answers 
given by parents, teachers, and the children themselves to 
certain questions in the Home Blank, School Blank, and In- 
terest Blank, and (2) records kept by the children of the 
books which they read during a period of two months. Con- 
siderable information was obtained for unselected children, 
especially by means of reading records. The reading of any 
group of children is so widely scattered that several thou- 
sand cases are usually necessary to establish many reliable 
differences in reading preferences. As our groups are rela- 
tively small, it is possible in most cases only to indicate gen- 
eral trends. With regard to the amount and general type of 
reading, however, the results are fairly conclusive. An ex- 
tended treatment of the data elsewhere makes it unneces- 
sary to give here more than a brief summary. 2 

AMOUNT OF READING AS ESTIMATED BY PARENTS 

In the Home Blank parents were asked to indicate the 
kind and amount of home reading at the following ages: 
before 5, 5 and 6, 7 and 8, 9 and 10, 11 and 12, 13 or above. It 
was specified that the amount was to be given in terms of 
hours per week, and that reading on school work should not 
be included. Probably because of the difficulty of answering 
the question, about 30 per cent of parents failed to respond 
and some who did respond gave data for only a part of the 
ages. Allowance must be made for a large amount of error 
in such reports. Children reported as having been accus- 
tomed to read 12 hours, 18 hours, or 25 hours a week at the 
age 6 years may have read more or less than these amounts, 
but the fact that parents report many such histories is at 
least interesting evidence. The figures reported are summa- 
rized in Table 167. 

a Written with the assistance of Margaret Lima. 

2 Lewis M. Terman and Margaret Lima: Children's Reading; A Guide 
for Parents and Teachers. Applet on, 1925. 

441 



442 TRAITS OF GIFTED CHILDREN 

TABLE 167 
HOURS OF READING PER WEEK AT VARIOUS AGES (PARENTS' ESTIMATES) 

Hours Before 5 5 and 6 7 and 8 9 and 10 11 and 12 13 or over 

Weekly Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls 






279 


207 


174 


118 


23 


20 














1 


6 


3 


7 


7 


2 


8 


2 


1 




2 






2 


16 


19 


15 


9 


9 


11 


5 


10 


3 


5 


1 




3 


11 


18 


24 


21 


17 


35 


9 


7 




9 




1 


4 


6 


5 


16 


35 


22 


34 


22 


23 


6 


8 






5 


4 




25 


24 


11 


26 


17 


8 


9 


6 


5 




6 


4 


2 


14 


13 


15 


25 


12 


20 


5 


5 


3 




7 


13 


7 


14 


13 


26 


17 


11 


13 


3 


6 




1 


8 


3 




6 


2 


19 


13 


17 


11 


3 


2 




1 


9 






2 




4 


8 


10 


3 


2 


10 




1 


10 




4 


12 


7 


33 


19 


19 


18 


11 


1 


10 




11 


1 




3 


1 


1 


5 


3 




1 


5 






12 




2 




2 


1 


7 


10 


6 


1 


1 




2 


13 










1 


1 








1 






14 






7 


4 


10 


10 


17 


6 


4 


1 






15 






2 


1 


2 


3 


8 


7 


5 


1 


2 




16 














1 


1 


3 


1 






17 














1 






2 






18 


1 




3 




3 






1 


3 








19 




















4 






20 


2 


2 


1 




9 


2 


8 


5 


5 


3 




2 


21 








1 


3 


2 




3 






3 




22 












1 














23 














1 






2 






24 
















1 




1 






25 or over 


1 




2 


1 




2 


7 


1 


2 


1 


1 


1 


Total 


























Reports 


348 


269 


327 


259 


210 


249 


183 


145 


66 


77 


25 


9 


Median 











1.5 


6.73 


5.13 


8.35 


6.77 


9.68 


8.08 


10.5 


11.75 


Mean 


1.06 


1.04 


2.90 


2.9 


7.2 


6.16 


9.6 


8.29 


10.44 


9.97 


12.5 


12.9 


S.D. 


3.06 


2.6 


4.32 


3.8 


5.5 . 


4.7 


5.8 


5.6 


6.0 


6,9 


7.3 


6.8 



According to parents' reports a good many gifted chil- 
dren read as much as seven hours a week before the age of 5 
years; by age 7 the average amount is six or seven hours a 
week and by age 13, twelve- hours. The amount reported for 
boys is slightly higher than that for girls from 7 to 12 years. 
The standard deviations are, however, extremely high. In 
cases where the amount of reading seemed impossibly high, 
all the available information regarding the child was sifted. 
For example, one boy was reported as reading twenty-five 
hours a week before the age of 5. Other information in our 
files showed that this boy learned to read before the age of 
3 years and that before 5 he was reading almost anything 
he could lay hands on. In every case of this kind the sup- 
plementary information lent credence to the parent's report. 



READING INTERESTS 443 

AMOUNT OF READING AS ESTIMATED BY TEACHERS 

In the School Blank teachers were asked to answer the 
following question : As compared with the average child of 
the same age, does this child read (1) very much, (2) more 
than average, (3) average amount, (4) less than average, 
(5) very little? In this case comparative data are available 
for the Control Group A. (See pages 177-178 for description 
of this control group.) Table 168 summarized the responses 
to above question. Ages were first tabulated separately, but 
since the comparison took age into account, the figures re- 
mained almost constant from age to age and it is necessary 
to give here only the results for the ages combined. As the 
control group did not go much below age 8 years, or the 
gifted group much above 13 years, the figures for each group 
are here presented only for children of ages 8 to 13. This 
gives a total of 429 gifted and 401 control children. 

TABLE 168 
TEACHERS' ESTIMATES OF AMOUNT OF READING 







Gifted 


Gifted 


Total 


Control 


Control 


Total 






Boys 


Girls 


Gifted 


Boys 


Girls 


Control 


1. 


Very much 


132 


99 


231 


19 


27 


46 


2. 


More than average 


86 


61 


147 


51 


37 


88 


3. 


Average amount 


24 


27 


51 


88 


89 


177 


4. 


Less than average 











31 


24 


55 


5. 


Very little 











22 


11 


33 


Total cases 


242 


187 


429 


211 


188 


399 




Mean amount 


1.55 


1.61 


1.58 


2.94 


2.76 


2.85 




Standard deviation 


0.67 


0.73 


0.69 


1.08 


1.04 


1.06 




Read more than 
















"average" 


90% 


86% 


88% 


33% 


34% 


34% 




Read less than 
















"average" 


0% 


0% 


0% 


25% 


19% 


22% 



The differences between the gifted and the control group 
are very striking. Of the 429 gifted, 88 per cent are rated as 
reading more than the average, while the distribution of 
ratings for the control group is fairly symmetrical around 
"average" as the mode. The sex differences are too small to 
be significant. 



444 TRAITS OF GIFTED CHILDREN 

RECORDS OF Two MONTHS READING 

The most reliable data on amount of reading were fur- 
nished by the children themselves in the form of records 
which they were asked to keep of the books read over a 
period of two months. Similar records were kept by a con- 
trol group. (Control Group C.) Each child was given a 32- 
page record booklet, size 3x5 inches. The first three pages 
were as follows : 

Reading Record 

Name of pupil Age 

Name of parent 



Date when record was begun 

Month Day Year 

Date when record was finished 



Month Day Year 

As soon as record is finished, mail this book to Professor Lewis M. Terman, 
Stanford University, California. 

(Read the instructions on the inside carefully.) 

WHY YOU ARE ASKED TO KEEP THIS RECORD 

I wish to find out what books children of each age like best, and in order 
to get the facts I am asking several hundred boys and girls to help me by 
keeping a record in this note book of all the books they read during a 
period of two months. 

When the note books have been returned to me I shall then be able to 
prepare and publish a "List of Best-Liked Books," which will be of great 
help to parents and teachers in selecting the books children of each age 
really enjoy most. By keeping this record you will therefore be doing 
something that will help to make the lives of thousands of children happier. 

LEWIS M. TERMAN. 

HOW TO KEEP THE RECORD 

1. For two months make a record of all the books you read, but do not 
include your regular school textbooks or books that some one else read to 
you. 

2. While keeping the record, you should read just your usual amount. 
Do not make a special effort to turn in a long list. For the present purpose 
it does not matter whether you read many or few. 

3. Make your record for each book on the day you finish reading it. Do 
not wait till the end of the week or the end of the month, for you might 
not then be able to remember all. 

4. Make your records neat and answer all the questions about each book. 

5. When you have recorded your reading for two months, mail this note 
book to Professor Lewis M. Terman, Stanford University, California. 



READING INTERESTS 445 

Then followed 29 pages, each as follows : 

1. Title of book 

Name of author 

Date -when you finished it, if you did finish it: Month Day 

If you did not finish it, tell why 



Below, make a cross before the statement that tells how well you liked it. 
(Give your real opinion, no difference what others think about the hook.) 

"One of the best I ever read/' 

"Liked it very well, better than most books." 

"Liked it fairly well." 

"Did not care much for it." 

"Did not like it at all." 

Had you ever read this book before? 

How many times before this time? 



Do you think you will want to read it again ?_ 



The booklets were distributed to the gifted group in May, 
1922. In most cases the first month of the record came 
within the school session, the second month within the sum- 
mer vacation. Many parents, when returning the booklets, 
wrote that the amount read was less than usual during the 
vacation, but it is possible that with others the reverse was 
the case. 

It was not originally planned to secure reading records 
for a control group, and when this was later decided upon it 
was not possible to begin the records until after the opening 
of the following school year. Booklets were distributed to 
Control Group C about October 1st, and the records cover 
in most cases the months of October and November. We do 
not know to what extent this discrepancy in time affected 
the results, or in what direction. One might suppose, how- 
ever, that it would favor the control group, as the gifted 
children received their booklets at a time when the pres- 
sure of school work is likely to be greatest. 

Control Group C, for which reading records were ob- 
tained, was a special group not used in any other connec- 
tion. It consisted of about 1,000 children from grades 2 to 8, 
inclusive, in Palo Alto, Redwood City, and Menlo Park, 
California. In the case of the control group the booklets 
were distributed and collected through the teachers. Each 
teacher assured her pupils that she would not examine their 
records and urged them to record all the books they read, 



446 TRAITS OF GIFTED CHILDREN 

with the exception of school texts. It is not likely, of course, 
that the children recorded their clandestine reading. If there 
was any difference between the gifted and control group in 
this respect it is probable that the reports of the gifted were 
less complete, as no effort was made to prevent parents of 
the gifted from inspecting the records of their children. 

Of approximately 600 children in the main experimental 
group who were asked to cooperate, 511 returned the record 
booklets in time for tabulation of results. The control group 
of approximately 1,000 children supplied 808 records. Table 
169 gives the average number of books read by both groups 
at ages 6 and 7 combined, 8 and 9, 10 and 11, etc. 

TABLE 169 

AVERAGE NUMBER OF BOOKS READ IN Two MONTHS, GIFTED AND 
CONTROL GROUPS 



Boys 
Gifted Control 


Girls 
Gifted Control 






Av. No. 




Av. No. 




Av. No. 




Av. No. 


Ages 


Cases 


Books 


Gases 


Books 


Cases 


Books 


Cases 


Books 


6 and 7 


16 


9.3 


17 





13 


11.5 


15 





8 and 9 


70 


12.4 


90 


4.6 


72 


15,2 


73 


5.6 


10 and 11 


123 


14.7 


139 


7.3 


92 


15.6 


147 


8.7 


12 and 13 


63 


12.8 


101 


5.8 


62 


16.5 


129 


7.7 


14 and 15 






56 


7.5 






41 


9.3 



Total 272 403 

Table 169 shows a large difference between the two 
groups. The gifted child of 7 years reads on the average 
more books than the unselected child reads at any age up 
to 15. One cannot say how many more pages they read, for 
the books they read are probably shorter than those read 
by older children. Gifted children of 8 and 9 read nearly 
three times as many books as unselected children of the 
same age, and above 10 approximately twice as many. By 
the age of 8-9 years the gifted have almost attained their 
maximum. 

In both groups there is a significant sex difference. Girls 
surpass boys by the following per cents in the number of 
books read at the different ages : 

Gifted Group Control Group 

8-9 23% 22% 

10-11 6% 19% 

12-13 29% 32% 

14-15 24% 



READING INTERESTS 447 

It is a little surprising to find that in both groups boys 
reach their maximum earlier than girls. This may be due to 
the fact that older boys are more likely to develop hobbies, 
such as outdoor sports, collections, mechanical pursuits, etc. 
Girls, lacking these diversions, naturally turn to books. 

An interesting fact not shown in Table 169 is that so 
many of the control group read no books at all during the 
two-months period. Of those 8 years old or older for whom 
booklets were returned, 13 per cent reported that no books 
were read. The proportion was doubtless much higher 
among the 20 per cent whose booklets were not returned. In 
the gifted group not a single child of 8 years or older who 
returned the booklet, reported fewer than two books read. 

INFLUENCE OF INTELLIGENCE ON TYPE AND RANGE OF READING 

General intelligence influences not only the amount of 
reading but also its quality and range. Comparative study of 
the two-months reading records of our gifted and control 
groups shows that the gifted read over a considerably wider 
range and that they read far more non-fiction and informa- 
tional material. On the whole, however, the most striking 
contrast is less in the type of books read than in the age at 
which they were read. A book that is well liked by an aver- 
age child of 11 or 12 is often read with enjoyment by the 
gifted child of 8 or 9. Table 170 shows for boys and girls 
of each group the per cent of books read belonging to vari- 
ous types. 

It will be noted that Table 170 tells nothing about 
the relative amount of reading done by the gifted as com- 
pared with unselected children. It tells merely what pro- 
portion of the reading actually done by each group belongs 
to each type of literature. For example, of the books read by 
the gifted boys, 8 per cent belong to the class "fairy stories, 
folk tales, and legends"; 9 per cent to "nature and animal 
stories," etc. 

Certain differences stand out very clearly. A larger pro- 
portion of the books read by gifted boys than of those read 
by control boys fall in the field of science, history, biography, 
travel, folk tales, nature and animal stories, informational 
fiction, poetry, drama, and encyclopedias; a smaller propor- 



448 TRAITS OF GIFTED CHILDREN 

TABLE 170 
CLASSIFICATION OF BOOKS READ BY GIFTED AND CONTROL GROUPS 

Per Cent of Books Read Belonging to Each Type 
Gifted Control Gifted Control 
Boys Boys Girls Girls 

Fairy tales, folk tales and legends 8% 5% 12% 5% 

Nature and animal stories 9% 4% 6% 6% 

History, biography, and travel 6% 3% 6% 2% 

Science 4% 0.5% 1% 0.2% 
Stories of adventure or mystery 
(mostly boys' juveniles and 

series books) 49% 63% 15% 22% 
Stories of borne and school 

life (mostly girls' books) 2% 3% 31% 33% 
Poetry and drama 1% 0.1% 1% 0.1% 
Children's encyclopedias 1% 0.1% 1% 0.1% 
Informational fiction, in- 
cluding the classics 19% 11% 14% 9% 
Emotional fiction (the popular 

novel and love story) 1% 6% 11% 19% 

tion in the fields of emotional fiction and stories of adventure 
and mystery. Almost exactly the same differences are found 
between gifted and control girls, except that here the two 
intelligence groups show about the same degree of prefer- 
ence for nature and animal stories. The differences indicate 
that the reading of the gifted is of a better average quality 
than that of the control group. 

SEX DIFFERENCES IN READING INTERESTS 

In the reading interests of very young children sex dif- 
ferences are not noticeable. By the age of 9 or 10 the boy 
begins to turn from fairy tales and fantastic stories to books 
of a more realistic nature, while the girl clings more to the 
imaginative story* By 11 or 12 the divergence is very 
marked, and the breach continues to widen up to adult life, 
when a certain amount of rapprochement takes place. 

Girls are more homogeneous with respect to reading 
tastes than are boys. Boys scatter their reading over a wider 
range. For example, Little Women is universally popular 
among girls, but we find no one book that has so wide a 
distribution among boys. This difference was also evident 
in the responses which a hundred graduate students, fifty 



READING INTERESTS 449 

men and fifty women, gave when asked to name the ten 
books read in childhood which most appealed to them. Of 
the women, 50 per cent listed Little Women, 36 per cent 
the Little Colonel books, 30 per cent Robinson Crusoe, and 
25 per cent Black Beauty. There were very few books not 
named by three or more women. The men's list showed 
variety. Robinson Crusoe, which appeared most frequently, 
was mentioned by only 13 per cent. Treasure Island came 
next with 12 per cent, and third, The Last of the Mohicans 
with 8 per cent. 

The narrower range of girls' interests is also indicated by 
their tendency to re-read books. In the reading record book- 
let the child was asked to state, regarding each book re- 
corded, whether he had ever read it before. The gifted and 
control children combined reported more than twelve thou- 
sand readings. Of the books recorded by girls, 30 per cent 
represented re-readings; of those recorded by boys, 18 per 
cent. 

Classifying the gifted and control by sex, we have the 
following classification of readings reported in the record 
booklets : 

TABLE 171 
CLASSIFICATION OF READING BY SEX 

Per Cent of Books Read Falling in Each Group 
Boys Girls 

Fairy tales, folk tales, and legends 1% 10% 

Nature and animal stories 7% 6% 

History, biography, and travel 5% 4% 

Science 3% 1% 
Stories of adventure or mystery 

(mostly boys' juveniles and series books) 56% 18% 
Stories of home and school life 

(mostly girls' books) 2.5% 32% 

Poetry and drama 0.4% 1% 

Children's encyclopedias 0.5% 0.4% 

Informational fiction, including the classics 15% 11% 
Emotional fiction (the popular 

novel and love story) 3.5% 16% 

The above figures reveal three outstanding contrasts. 
As compared with boys, girls read (relatively to total 
amount read) (1) more than twelve times as many books of 
home and school life; (2) nearly five times as much emo- 



450 TRAITS OF GIFTED CHILDREN 

tional fiction; and (3) only a third as many stories of ad- 
venture or mystery. Girls care more than boys for fairy 
stories, and boys more than girls for books of science, his- 
tory, biography, travel, and informational fiction. When 
romance enters into a boy's book it must be so intermingled 
with action that the sentiment is not too obtrusive. The 
strong human interest in girls is shown by their intense 
liking for books of the Little Women type. 

To what extent the sex differences in reading interests 
reflect differences in native endowment, and to what extent 
the subtle effects of social ideals and training, it is impossi- 
ble to say. It is the tradition of our race that men should be 
interested in such things as industries, machinery, and the 
sciences, and that woman's sphere is the home. The girl is 
exposed to this tradition from her earliest years, and it 
would be surprising if such long-continued and pervading 
suggestion did not leave its mark on her reading interests. 

It is worthy of note, however, that although boys show 
practically no interest in girls' books, girls show a most de- 
cided interest in boys' books. Girls read with interest Treas- 
ure Island, The Call of the Wild, and other popularly ac- 
cepted boys' books. They read the Boy Scout books and 
other boy adventure series. Few boys, however, read Little 
Women or Rebecca of Sunnybrook Farm, and they rarely 
open a girls' book on stories of school life. From the reading 
records of our children it was found that 18 per cent of the 
girls' reading was in the field of boys' books, but only 2 per 
cent of the boys' reading was the human interest story of 
home or school life that girls so much enjoy. 

PHEFERENCE RATINGS OF BOOKS READ 

It will be recalled that the record booklet requested the 
child to rate each book read by checking one of the follow- 
ing statements : 

One of the best I ever read. 

Liked it very well, better than most books. 

Liked it fairly well. 

..Did not care much for it. 

Did not like it at alL 

This request was complied with in nearly all cases. It 
was hoped that the results would be worth summarizing for 



READING INTERESTS 451 

the individual books recorded, but, as it turned out, the read- 
ing covered such a wide range that the number of ratings for 
a given book was in most cases too small to warrant the com- 
putation of average ratings for individual books. However, 
ratings were distributed and averaged for the books of each 
general class. The averages are shown in Table 172 for the 
gifted and control groups. In computing averages the five 
degrees of preference were assigned the values 1, 2, 3, 4, 5, 
in order from greatest to least liking. 

TABLE 172 
MEAN PREFERENCE RATINGS OF BOOKS READ 

Gifted Gifted Control Control 
Boys Girls Boys Girls 

Fairy tales, folk tales, and legends 3.2 2.6 3.0 2.7 

Nature and animal stories 2.7 2.4 2.8 2.6 

History, biography, and travel 1.8 2.0 2.5 2.8 

Science 2.4 2.7 3.3 3.6 
Stories of adventure or mystery 

(mostly boys* juveniles and 

series books) 1.5 1.8 1.3 2.0 
Stories of home and school 

life (mostly girls' books) 3.4 1.3 3.2 1.4 

Poetry and drama 3.9 3.5 4.3 3.9 

Children's encyclopedias 2.9 3.2 3.7 4,3 
Informational fiction, 

including the classics 1.9 1.8 2.2 2.1 
Emotional fiction (the popular 

novel and love story) 2.3 1.5 2.5 1.8 

Table 172 shows that in general the books which are 
most frequently read are the books which are best liked. 
This was to be expected, but the figures lend emphasis. The 
average of the rating is decidedly higher for the gifted than 
for the control, and slightly higher for girls than for boys. 
This may be taken as additional evidence that reading is 
more enjoyed by gifted than by unselected children, and 
more enjoyed by girls than by boys. 

FAVORITE BOOKS 

In the Interest Blank the children were asked to name 
four or five books they had most enjoyed reading in the last 
year. Responses were obtained from 602 of the main experi- 
mental group and from 1,225 unselected school children. The 



452 TRAITS OF GIFTED CHILDREN 

books listed cover so wide a range that the number who 
name any one book is too small to give a reliable compari- 
son of the preferences of the two intelligence groups. The 
sex differences, however, were more marked than the intel- 
ligence differences, and the two intelligence groups have 
therefore be'en combined to secure a list of the twenty books 
most liked by boys and the twenty most liked by girls. They 
are as follows, listed in rank order: 

The Twenty Books Most Liked by Boys 

1. Treasure Island* Stevenson 

2. Call of the Wild* Jack London 

3. Tom Sawyer Mark Twain 

4. Eobinson Crusoe Defoe 

5. Three Musketeers* Dumas 

6. Ivanhoe* Scott 

7. Huckleberry Finn Mark Twain 

8. Penrod Tarkington 

9. Sherlock Holmes Conan Doyle 

10. Kidnapped Stevenson 

11. Black Beauty Sewall 

12. Swiss Family Robinson Wyss 

13. Connecticut Yankee - Twain 

14. Tale of Two Cities* Dickens 

15. Count of Monte Cristo Dumas 

16. Penrod and Sam Tarkington 

17. White Fang Jack London 

18. Last of the Mohicans Cooper 

19. Jungle Books Kipling 

20. Oliver Twist Dickens 

The Twenty Books Most Liked by Girls 

1. Little Women Alcott 

2. Anne of Green Gables Montgomery 

3. Ivanhoe* Scott 

4. Little Men Alcott 

5. Treasure Island* Stevenson 

6. Laddie Porter, G. S. 

7. Three Musketeers* Dumas 

8. Alice in Wonderland Carroll 

9. Heidi Spyri 

10. Pollyanna Porter, E. 

11. Secret Garden Burnett 

12. Rebecca of Sunnybrook Farm Wiggin 

13. David Copper field Dickens 

14. Little Lord Fauntleroy Burnett 

15. Call of the Wild* London 

16. Eight Cousins Alcott 



READING INTERESTS 453 

17. Freckles Porter, G. 

18. Little Minister Barrie 

19. Tale of Two Cities* Dickens 

20. Uncle Tom's Cabin Stowe 

Series books could not be included in the above lists, 
since they were usually mentioned as a series rather than 
as individual books, but when they were treated separately 
it was found that for girls the Oz books were the most popu- 
lar series, with the Little Colonel books coming next. For 
boys, the Book of Knowledge showed a surprising lead, with 
the Oz books second. 

Some of the interesting facts brought out by the above 
table are the following: 

1. With the exception of the Book of Knowledge, all of 
the most liked books are fiction. This is partly explained by 
the fact that the non-fiction reading covered such a wide 
range that agreement of choice occurred but seldom; never- 
theless, even if liberal allowance is made for this factor, it 
appears that fiction still holds first place in the reading pref- 
erences of both boys and girls. 

2. In the type of fiction preferred, striking sex differences 
are seen. These are in agreement with the findings already 
set forth; the boys prefer stories of adventure and mystery, 
while the girls prefer stories of home and school life. 

3. Only five titles (those starred) appear in both lists; in 
other words, the lists are mutually exclusive to the extent of 
75 per cent. That the lists overlap at all is due almost en- 
tirely to the fact that girls frequently read boys' books. Dis- 
tinctly girls' books are rarely read by boys. The five titles 
appearing in both lists are Treasure Island, Call of the Wild, 
Ivanhoe, Three Musketeers, and Tale of Two Cities. 

SUMMARY 

1. The mean number of hours of reading per week by 
gifted children (parent's estimates) increases from about 6 
at age 7 to 12 at age 13. At all ages below 13 the mean is 
slightly higher for gifted boys than for gifted girls. 

2. According to teachers' estimates, 88 per cent of the 
gifted and 34 per cent of the control group read more than 
the average child; per cent of the gifted and 22 per cent of 
the control group read less than the average child. 



454 TRAITS OF GIFTED CHILDREN 

3. Children of the gifted groups and of a control group 
kept a record of the books read during two months. Analy- 
sis has been made of such records for 511 gifted and 808 con- 
trol children. 

4. The reading records show that the average gifted child 
of 7 years reads more books in the two months than the aver- 
age of the control group for any age up to 15. The average of 
gifted children at 8 or 9 years is three times that for the con- 
trol group of the same age. By this time the average gifted 
child has almost attained his maximum as to number of 
books read. 

5. Of the control group who had attained the age of 8 
years, 13 per cent read no books at all during the two 
months; of the gifted group who were 8 years old or older, 
none had so poor a record. 

6. Classification of the books read by the two groups 
brought out the fact that the gifted children read over a con- 
siderably wider range than the control children, and that 
they read more science, history, biography, travel, folk tales, 
informational fiction, poetry, and drama. On the other hand, 
in proportion to the total number of books read the gifted 
read fewer books of adventure or mystery and far less emo- 
tional fiction. 

7. In both gifted and control groups the boys scatter 
their reading over a much wider range than do the girls. 
Girls are much more likely than boys to read a book two or 
more times. 

8. Boys read about three times as many books of adven- 
ture or mystery as the girls read of this type, while the girls 
read nearly five times as much emotional fiction as boys 
read. 

9. When the twenty most popular books were listed for 
each sex only five books were found in both lists. 

10. With the exception of the Book of Knowledge, all of 
the twenty best-liked books (both lists) were fiction. 



CHAPTER XVI 

TESTS OF INTELLECTUAL, SOCIAL, AND 
ACTIVITY INTERESTS 

Jennie Benson Wyman 

Introductory statement. Apart from the data of the two preced- 
ing chapters, more or less light has been thrown on the interests of 
gifted children in a numher of divisions of this study. The educa- 
tional history of the children, as given hy parents and teachers 
(Chapter X), the tests of information and achievement (Chapters 
XI and XII), the ratings given by parents and teachers on per- 
sonality traits (Chapter XVIII), and the comparative study of the 
mental development of eminent individuals (Volume II), have all 
suggested the importance of interest as a determinant of achieve- 
ment and as a factor to be considered in individual diagnosis and 
educational treatment. In many cases, however, the evidence has 
been indirect and inferential. It is inadequate for its purpose in the 
same way that class marks and teachers* estimates of the quality of 
school work are inadequate to establish the exact status of a child 
with respect to his educational accomplishment. In both cases the 
data supplied by parents, teachers, and others are open to suspicion 
on the ground that the judgments rendered may have been influenced 
by suggestion, "halo" effects, or other forms of bias. We do not 
know in what respects and to what extent the various kinds of testi- 
mony and ratings would have differed from those obtained if parents 
and teachers had been led to believe that the tests showed the chil- 
dren to be average or inferior in intelligence instead of "gifted." 
Even the testimony of the children themselves in regard to their 
interests cannot be accepted at its face value. In the first place, 
children have no standard by which to judge the absolute strength 
of their interests. In the second place, they are likely to be misled 
in regard to the relative amount of interest they have for the different 
studies and activities by the influence of associations, like or dislike 
for a teacher, etc. The need, of course, is for a measure of interest 
that would be as objective, as consistent, and as valid as the best 
measures of intelligence or educational achievement. 

It is a large undertaking, however, to set about the derivation of 
such a test. The quantitative study of interest is not only a relatively 
new field, but one which is also inherently difficult. Interest seems 
very intangible in comparison, for example, with intelligence. Intel- 
ligence is abiding, it does not ordinarily take wings, or completely 
alter its form and substance, as interest is so likely to do, the moment 
one approaches it with a test. Obviously the measurement of interest 
is a problem which cannot be solved by any single attack, however 

455 



456 TRAITS OF GIFTED CHILDREN 

skillfully planned. The attempt, nevertheless, seemed worth making, 
and the task of devising a valid method was entrusted to Mrs. Jennie 
Benson Wyman, a graduate student in education and psychology at 
Stanford University. Mrs. Wyman devoted to the problem the greater 
part of her time for more than two years, working chiefly under the 
direction of Dr. Kelley. 

It is possible to present here only a brief description of her 
methods and a summary of the most important of her findings. 

LEWIS M. TERMAN. 

METHOD OF APPROACH 

A canvass of the experimental studies of emotional and 
personality traits led to the belief that the free association 
method offered the most promising line of attack for the 
measurement of interest. The investigations of Jung, of 
Kent and Rosanoff, and of many other workers had shown 
clearly that an individual's characteristic tendencies of 
thought, emotion, and will may reveal themselves in the ap- 
parently trivial responses to stimulus words in such a test. 
If it is possible thus to uncover mental complexes which lead 
to abnormal behavior, and this possibility has been suffi- 
ciently demonstrated, why should not important underlying 
interest trends also be brought to light by an association test 
specially planned for the purpose? 

It was necessary, of course, to limit the undertaking to 
the measurement of a few significant aspects of interest. An 
extended consideration of the types of interest which are 
probably most influential in molding one's personality, de- 
termining one's attitudes, and shaping one's life habits led to 
the selection of the following three types for investigation : 
(1) intellectual interests, (2) social interests, and (3) activity 
interests. These seemed to be the most promising from the 
point of view of experimental procedure, and at the same 
time highly significant in connection with the study of gifted 
children. According to the traditional psychology of genius, 
the gifted individual tends to develop his intellectual inter- 
ests at the expense of his interests in social and activity 
fields. The social and activity interests, perhaps in the first 
place inherently weak, are supposed to undergo continuous 
atrophy until the individual withdraws from active partici- 
pation in affairs and becomes socially disinterested and in- 
effective. The ill-balanced Hamlet is often pointed to as an 
example of what results from too much thinking combined 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 457 

with too little "mixing" and doing. At any rate, it would be 
extremely desirable to have for each of our gifted children 
a set of objectively derived scores which would measure 
both the relative and the absolute strength of the intellectual, 
social, and activity interests. 

In order to test these three aspects of interests the dic- 
tionary was scanned for words which, when used as stimulus 
words in a free association test, were equally adapted to 
provoke responses due to intellectual interests, social inter- 
ests, or activity interests. That is, with respect to these in- 
terests the stimulus words were to be of the "balanced" type, 
so that a small preponderance of interest in any one of the 
three directions would reveal itself in the nature of the re- 
sponses. First a list of 800 words was drawn up. For a pre- 
liminary experiment this was reduced to 80, composing four 
comparable lists of 20 words each. It was necessary that the 
words retained should satisfy the following conditions: 

1. They should be in common use. All were excluded 
which were not found in first 2,000 of Thorndike's word list. 

2. They should not be too readily definable. 

3. They should not be too much "tied" with other words 
through continued association with them. 

4. They should not be words which would yield a large 
percentage of opposites in the responses. 

5. They should be words which would not leave too strong 
an impression on the mind and produce a carry-over effect. 

6. The list as a whole should have a cyclic character so 
that it could be broken up into four comparable brief lists. 

The 80 words thus selected were tried out on a number 
of subjects individually and in small groups, in order to get 
an idea of the technique that should be used with respect to 
presentation of stimulus, method of response, timing, etc. 
Group administration was found to be feasible. The stimu- 
lus words were presented visually and at the same time pro- 
nounced by the experimenter. Response was by writing. A 
rate of one stimulus word every eight seconds was found 
the most satisfactory for children of the upper elementary 
grades. 

PRELIMINARY EXPERIMENT 

The provisional list of words was given to 175 pupils of 
the sixth and seventh grades, 40 words on one day and the 



458 TRAITS OF GIFTED CHILDREN 

other 40 two weeks later. The children were simply told to 
"write down in one word" what the word shown made them 
think of. No examples of how to respond were given. 

Responses were then analyzed for their significance. It 
was necessary to score each response three times; first for 
its intellectual value, next for its social value, and finally 
for its activity value. Rating was at first on a scale of 5 to 
for each response and was based on the experimenter's sub- 
jective judgment as to the intellectual, social, or activity value 
of the response. The following reliability coefficients and 
intercorrelations were found for two groups of 12 year olds 
selected from the 175 children tested: 

Reliability coefficients (SO word list) : 

City A (51 cases) City B (39 cases) 

Intellectual interest .85 .89 

Social interest .66 .66 

Activity interest .48 .45 

Intercorrelations 

Intellectual vs. social .36 .56 

Intellectual vs. activity .20 .10 

Social vs. activity -.08 .21 

Eighteen of these children were given a rank order rat- 
ing by one teacher on each of the three interest traits. This 
gave the following rank-order correlations between test 
score and teacher's estimate: 

Intellectual interest .67 

Social interest .49 

Activity interest .22 

The above results were deemed promising, but it was 
found necessary to make extensive revisions in the list of 
stimulus words and to derive a more objective method of 
scoring. Some of the words gave little variety of response, 
some tended to provoke too many definitions, and some of 
the pairs that were believed to be comparable proved not 
to be. Of the original list of 80 words, 37 were eliminated, 
and enough new words were added to give two comparable 
lists of 80 each. 

The new lists were then given to the children who had 
taken the provisional list and to several other groups. As 
there were 47 words common to the provisional list and the 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 459 

revised list, it was possible to secure evidence as to the con- 
sistency of a test of this nature when it is repeated. In this 
case 13 months had elapsed since the first test was given. In 
order to make a valid comparison of the results of the two 
tests it was of course necessary to have a strictly objective 
method of scoring. For the immediate purpose of the com- 
parison the following classification of responses was used: 

1. Identity. Synonyms. 

2. Contrast. 

3. Causal Dependence. 

4. Co-adjunction. 

5. Subordination. 

6. Supra-ordination. 

7. Definitions or explanations. 

8. Co-ordinations of undetermined quality. 

9. Co-existence. 

10. Judgment of value (predicate). 

11. Judgment of fact (predicate). 

12. Subject and object relation (predicate). 

13. Associations denoting time, place, means, and pur- 

pose. 

14. Reaction omitted. 

15. Egocentric reaction. 

16. Word of phrase completion current phrases. 

17. Repetition of stimulus word. 

18. Meaningless reaction. 

19. Non-specific reaction. 

20. Clang-reactions. 

There were 53 children who took both tests. Their re- 
sponses were in each case classified into the above 20 cate- 
gories, and a table for calculating the Mean Square Contin- 
gency was drawn up. Correction for too fine grouping 
necessitated grouping categories 2, 4, 6, and 8 together, and 
17, 18, 19, and 20 together, which reduced the number of 
categories to 14. The coefficient for Mean Square Contin- 
gency was calculated and found to be .856.0056. There is 
accordingly a rather remarkable degree of constancy in the 
responses given by a child from year to year. It may be 
held, however, that this correlation is spuriously high for 
two reasons: (1) because of the memory factor; and (2) be- 
cause certain responses are so "tied" with certain stimulus 



460 



TRAITS OF GIFTED CHILDREN 



words that they appear almost inevitably in the responses of 
any group of subjects. To meet these objections, the re- 
sponses were examined again and all those which were 
given in the identical word in the two tests were eliminated. 
This necessitated the elimination of 22 per cent of the re- 
sponses. The responses remaining yielded a Mean Square 
Contingency coefficient of .80. This showed that the func- 
tions measured by the test are not evanescent, but to a sur- 
prising degree permanent. 

Further analysis of the responses of 31 seventh grade 
children to the revised list of 160 words showed the necessity 
for further revision. A few of the words brought too many 
failures to respond and a few tended to provoke persevera- 
tion. After necessary eliminations enough words remained 
to yield two comparable lists of 60 each. These are as 
follows : 



1. summer 


31. evening 


1. night 


31. sundown 


2, easy 


32. hard 


2. simple 


32. difficult 


3. diamond 


33. ring 


3. gem 


33. dress 


4. tire 


34. play 


4. join 


34. enjoy 


5. dog 


35. learn 


5. control 


35. need 


6. fair 


36. band 


6. white 


36. music 


7. school 


37. dark 


7. college 


37. black 


8. help 


38. platform 


8. protect 


38. stage 


9. nature 


39. pity 


9. sky 


39. watch 


10. active 


40. thrill 


10. restless 


40. excite 


11. dream 


41. idle 


11. wonder 


41. useful 


12. shock 


42. hero 


12. fault 


42. castle 


13. joy 


43. vacation 


13. pleasure 


43. holidays 


14. dislike 


44. master 


14. detest 


44, captain 


15. nut 


45. bat 


15. paper 


45. rod 


16. go 


46. fun 


16. travel 


46. mischief 


17. angel 


47. power 


17. princess 


47. rain 


18. nice 


48. interested 


18. alone 


48. interesting 


19. water 


49. fond 


19. current 


49. good 


20. boy 


50. trip 


20. girl 


50. journey 


21. wish 


51. make 


21. desire 


51. form 


22. museum 


52. yard 


22. history 


52. island 


23. delight (ed) 


53, aim 


23. contented 


53. try 


24. work 


54. fairy 


24. train 


54. giant 


25. cave 


55. exercise 


25. adventure 


55. game 


26. pleasant 


56. companion 


26. happy 


56. friend 


27. house 


57. career 


27. marble 


57. science 


28. imagine 


58. fire 


28. invent 


58. camp 


29. range 


59. like 


29. country 


59. prefer 


30. admire 


60. great 


30. attract 


60. grand 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 461 

An attempt was made to improve the scoring of responses 
for intellectual interest on the basis of estimated values as- 
signed by two judges to each of 26 classificatory categories, 
such as identity, contrast, causal dependence, co-adjunction, 
subordination, supra-ordination, definition, etc. The corre- 
lation between the values assigned by the two judges was 
.849. When this method of scoring was tried out, it was 
found to yield lower coefficients of reliability and validity 
than the more subjective method used earlier. However, as 
some kind of objective scoring method was deemed neces- 
sary, it was decided to work out empirically one based upon 
a comparative study of the responses of specially selected 
groups. 

DERIVATION OF THE SCORING METHOD 

The final list of 120 words was given to 689 gifted chil- 
dren, most of whom belonged to the main experimental 
group, and to a control group consisting of 609 unselected 
children of grades 4 to 8 in San Jose. 1 In addition, it was 
given to two special groups, which may be designated as A 
and B. Special group A consisted of 115 sixth and seventh 
grade pupils, and was to serve for the final try-out of the 
new scoring method and for the statistical analysis of the 
test itself. Special group B consisted of 150 seventh grade 
pupils. This group was given the test in two parts of 60 
words each, with an interval of two weeks between the sit- 
tings. This group was reserved for the determination of the 
reliability of the test. 

The teachers of all the seventh grade classes tested in 
special groups A and B were requested to make out three 
separate rank orders of their pupils, one for intellectual in- 
terest, one for social interest, and one for activity interest. 
The meanings to be attached to these terms were explained 
to the teachers as follows : 

Intellectual Interest 

A person with a high degree of intellectual interest is one inter- 
ested in knowing interested in getting at the meaning of things 
the person who elects to "know," rather than to "do." 



group may be designated as Control Group D, to distinguish it 
from control groups A, B, and C mentioned in preceding chapters. 



462 TRAITS OF GIFTED CHILDREN 

Social Interest 

Interest in persons do not confuse social interest with social 
performance. The most popular person is not necessarily the one 
with the highest social interest. 

Activity Interest 

A person with a high degree of activity interest is the one who is 
interested in "doing" things the leader the one quick to respond 
but not necessarily the one fondest of outdoor games. A person 
who prefers to "take part" rather than to "watch." 

These types of interest are not mutually exclusive. 

In the case of the seventh grade pupils of the control 
group the teachers were not asked to prepare rank orders, 
but merely to designate 10 pupils with much and 10 with 
little intellectual interest, 10 with much and 10 with little 
social interest, and 10 with much and 10 with little activity 

interest. 

In one school there were 39 pupils for whom it was possi- 
ble to secure rank order ratings from four teachers, A, B, C 
and D. Intellectual interests were rated by A and B, social 
interests by A and C, and activity interests by A and D. The 
rank-order correlations yielded the following reliability co- 
efficients for the ratings : 

rAB=.507 for intellectual interest 
rAC=.694 for social interest 
rAD=.273 for activity interest 

Later, B, C, and D (A was no longer available) were 
asked to rate their pupils again; this time by stating merely 
whether a given pupil was intellectually interested or not, 
socially interested or not, interested in activity or not. 
This gave a second rating by B on intellectual interest, a 
second by C on social interest, and a second by D on activity 
interest. Bi-serial r's were then calculated as follows: 

For B's two ratings on intellectual interest, .93 .04 

" C*s " " " social " .77 .07 

" D's " " " activity " 65 .08 

On the basis of all the above ratings, six groups of chil- 
dren were made up, all in the seventh grade at the time the 
test was given, and all between the ages of lQ l /z and 14 
years. These groups were as follows : 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 463 

1. With intellectual interest* 69 

2. Without " " 58 

3. With social " 71 

4. Without " " 67 

5. With activity " 71 

6. Without " " 56 

* "With** and "without** are here used as comparative terms, "with" indicating 
the upper end of a continuous distribution and "without" the lower end. 

These groups were to serve as a basis for deriving the 
scores to be assigned to response words. A particular re- 
sponse word given by a large percentage of the "with'* 
group and by a small percentage of the "without" group (or 
vice versa) has a high differentiating value and is diagnostic. 
For each response, therefore, the frequency was found sepa- 
rately for the "with" group and the "without" group, in 
terms of the per cent who gave it. The difference between 
these two per cents was also found. Next the S.D. of each 
per cent was computed, and the S.D. of the difference be- 
tween the two per cents. Comparing the difference between 
the per cents with the S.D. of the difference gave a measure 
of the amount of a particular kind of interest involved in a 
particular response word. Accordingly, the score assigned to 
a response word was the difference between the per cents of 
"with" and "without" groups giving it divided by the S.D. of 
this difference. The ratios thus obtained were then trans- 
muted into a 0-20 scale, indicating no interest and 20 maxi- 
mum interest. It was necessary, of course, to carry through 
this procedure separately for the three kinds of interests. 
In all, there were 10,880 response words to evaluate, each 
three times. Upward of 13,000 additional responses were 
encountered in scoring the papers of other groups of chil- 
dren tested. 

A serious difficulty encountered in the evaluation of re- 
sponse words arose from the fact that sometimes a stimulus 
word brought two or more responses obviously differing 
little in significance but appearing with very different fre- 
quency. Moreover, in scoring the papers of the various 
groups tested, many response words were encountered which 
were not given by any pupil in the six "with" and "without" 
groups used in the derivation of the scoring method. For 
both of these reasons the grouping of response words be- 



464 



TRAITS OF GIFTED CHILDREN 



came necessary. 1 This involved a considerable amount of 
subjective judgment, the effect of which is to reduce to some 
extent the reliability and validity of the test. 

Each of the 10,880 response words encountered was thus 
allotted three scores, one for each of the three types of inter- 
est. The key for scoring the responses to a single stimulus 
word fills an entire page. Two sample pages, for the scoring 
of responses to the stimulus words gem and grand follow. 
The three columns of scores are, in order, those for intellec- 
tual interest, social interest, and activity interest. 



SCORES FOR RESPONSES TO 
quency Responses ISA 



44 

114 diamond 

79 stone 

57 jewel (s) 

S3 ruby 

14 ring(s) 

12 exercise (s) 

12 precious 

10 pearl 

10 razor 

6 boy 

basketball 

beautiful 

biscuits 

blade 

bread 



448 

20 11 15 

12 10 15 

12 13 5 

20 11 15 

9 13 5 

3 9 12 

10 11 8 
20 11 15 

396 

11 13 9 

8 9 12 
10 11 8 
3 9 12 
396 
3 9 12 



cake 


3 


9 


12 


cast 


8 


9 


6 


city 


3 


9 


12 


Columbia 


3 


9 


12 


Columbus 


3 


9 


12 


cook 


10 


13 


12 


country 


3 


9 


12 


diamond ring 


9 


13 


5 


doughnuts 


3 


9 


12 



easy 
eat 

emerald 
events 



10 11 8 

10 13 12 

20 11 15 

867 



STIMULUS WORD GEM 
Responses I 



garnet 

gold 

good 

green 

gum 

gun 

hard 



20 
10 
10 
10 
12 
10 



jelly 
jewelry 
Jim 
juice 

light 

meet 

memory 

mine 

mineral 

money 

muscle 

name 
nursery 

ocean 

pearl divers 
physical aid 
play 



3 
9 

10 
8 



8 
11 



s A 

11 15 

11 15 

11 8 

11 8 

10 12 

10 7 



10 11 8 



Island of Gems 10 10 10 



9 12 
13 5 



10 

9 



7 
6 



12 10 12 
867 
975 
12 10 15 
7 5 
9 12 



865 
3 13 12 

7 6 13 

20 11 9 

8 9 12 
10 10 12 



1 The method by which the grouping was done is too complicated to 
describe here. 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 465 



Responses 


I 


s 


A 


Responses 


I 


s 


A 


playground 


3 


13 


12 


sparkle 


10 


10 


12 


playroom 


3 


13 


12 


sparkling 


10 


11 


8 


precious stone 


12 


11 


15 


sticks 


10 


10 


10 


present 


7 


13 


5 










pretty 


10 


11 


8 


thing 


8 


6 


5 










topaz 


20 


11 


15 


razor blade 


3 


9 


6 


treasures 


8 


13 


5 


red 


10 


11 


8 


tree 


8 


9 


6 


rich 


10 


11 


8 










riches 


8 


7 


5 


valuable 


10 


11 


8 


round 


10 


11 


8 


















water 


7 


6 


13 


sapphire 


20 


11 


15 


wonderful 


10 


11 


8 


show 


7 


6 


13 











SCORES FOR RESPONSES TO STIMULUS WORD GRAND 



Fre- 


















quency Responses 


i 


s 


A 


Responses 


i 


s 


A 


52 




6 


8 


11 


delightful 


10 


12 


8 




73 


wonderful 


16 


13 


8 


dog 


10 


8 


6 


33 


good 


10 


7 


8 










31 


great 


10 


11 


8 


evening 


6 


10 


10 


27 


nice 


10 


13 


8 










25 
20 


beautiful 
fine 


12 
16 


12 

10 


8 
8 


feeling 


6 


10 


10 


13 


big 


8 


8 


8 










15 
12 
10 


piano 
lovely 
ball 


8 
10 
6 


9 
12 
11 


11 
8 
10 


game 
glorious 
glorious time 
grand 


6 
16 
13 
12 


8 
12 
10 
8 


10 
8 
10 
14 




admire 
army 
auto 
awful 


12 
6 
6 
12 


12 
9 
10 
10 


14 
6 
10 
8 


grandmother 
grandpa 
grand piano 
grandstand 
greatness 


12 
12 
12 
12 
13 


13 
13 
13 
13 
12 


14 
14 
14 
14 
10 




band 


8 


9 


11 












believe it 


10 


10 


12 


hair 


6 


8 


10 




bicycle 


6 


10 


11 


he 


8 


8 


12 












high 


8 


10 


8 




Cadillac 


14 


10 


11 


holidays 


13 


10 


10 




camp 


10 


12 


11 


home 


6 


6 


10 




candy 


10 


10 


11 


honour 


13 


12 


10 




Canyon 


6 


9 


11 


horrid 


12 


10 


8 




castle (s) 


14 


6 


10 


horse 


10 


8 


6 




church 


6 


6 


10 


hotel 


6 


6 


10 




clothes 


6 


8 


10 


house 


6 


6 


10 



466 



TRAITS OF GIFTED CHILDREN 



Responses 

imposing 
independent 

journey 
king 

lake 

large 

laugh 

machine 

magnificent 

majestic 

make 

man 

many 

marvellous 

millionaire 

mountains 

music 

night 
noble 

old 

opening 

opera 

palace 

party 

person 



i s 

16 12 
8 10 



13 



8 10 

9 6 



6 6 10 
8 12 6 
6 10 12 



6 10 10 

16 8 8 

16 10 8 

10 12 12 

6 8 

10 10 

16 10 

13 9 



14 6 10 

10 8 11 

6 10 10 

16 10 8 

8 10 8 

6 8 10 

6 9 11 

14 6 10 

6 10 10 
686 



Responses 

play 
pleasant 
pretty 
programme 

queens 
river 



ISA 

10 12 12 

10 11 8 

8 12 8 

6 8 11 

13 9 6 

6 6 10 



science 


14 


8 


10 


show 


16 


13 


11 


something nice 


10 


10 


8 


splendid 


16 


8 


8 


stand 


6 


9 


11 


stately 


16 


10 


8 


study 


10 


12 


12 


sublime 


16 


10 


8 


surprise 


14 


10 


14 


swell 


10 


11 


8 


swimming 


10 


13 


14 



theatre 

time 

trees 

useful 
Valentino 

wonder 
woods 



6 6 10 

6 8 10 

14 6 10 



8 10 
13 12 



8 



14 13 10 
14 6 10 



An examination of the above scores will bring many sur- 
prises. For example, one could not have foreseen that in 
response to the stimulus word grand, "fine" or "wonderful" 
would have a much higher intellectual value than "opera"; 
that "horse" would have a higher intellectual than activity 
value; that "nice" would have nearly twice as high social 
value as "good," etc. Nor could one have foreseen that in 
response to the stimulus word gem, "diamond" would have 
an intellectual value of 20, as compared with 9 for "jewelry"; 
or that "treasures" would have a higher social value than 
"money." One is not surprised to find that in response to 
most stimulus words "school" is social and "book" and 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 467 

"study" non-social, or that "help" is more social than "pity" ; 
but one could not have predicted with certainty that "girl" 
would prove to be more social than "boy," or "swimming" 
more active than "dancing." 

Doubtless if the test were given to other criterion groups 
of children similarly chosen, and if the resulting data were 
similarly treated, the values found for the various responses 
would differ more or less from those which have been ob- 
tained. For differentiating between the criterion groups 
which were used, the values given are approximately correct. 
If one would learn the best method of scoring a mental test 
it is necessary to disregard all a priori considerations and 
follow the empirical method. For example, it was demon- 
strated empirically that failure to respond should not be 
given a score of zero, but a fairly high positive score, dif- 
fering, however, with the different stimulus words. 

Comparison of scores of unselected and gifted groups 
suggests that a test of this kind could be devised which would 
have considerable value as a measure of intelligence. Some 
of the words in the present list could be used for this pur- 
pose. In response to the stimulus word night, for example, 
33 per cent of the gifted and 7 per cent of the control chil- 
dren give "darkness," while 33 per cent of the gifted and 72 
per cent of the control give "dark." In response to the stimu- 
lus word house, 40 per cent of the gifted and 15 per cent of 
the control give "home"; while 2 per cent of the gifted and 
20 per cent of the normal give "wood." Many other cases of 
this kind were found, but no attempt was made to derive a 
measure of intelligence from the list used. 

RELIABILITY OF THE TEST 

Reliability coefficients, S.D's, intercorrelations, and P.E's 
of scores were computed for the final list for unselected 12 
year old children. These are given in Table 173. 

The reliability coefficients are surprisingly high, consid- 
ering that the entire test of 120 words may be given in about 
16 minutes. The probable errors of the scores are correspond- 
ingly small. By trial it was found that when the test was 
given in two sittings separated by 10 days, instead of in a 
single sitting, the reliability coefficient for intellectual inter- 



468 TRAITS OF GIFTED CHILDREN 

TABLE 173 
RELIABILITY COEFFICIENTS AND INTERCORRELATIONS FOR FINAL LIST 



(1) Reliability 
Coefficients 



(2) Standard 
Deviations 



(3) Inter- 
correlations 



(4) P. E. of an 
Individual 
Score 

est was very much lowered, and that for activity interest 
considerably lowered. With social interest the difference 
was not significant. 

VALIDITY OF THE TEST 

The procedure adopted in determining the proper scor- 
ing responses guarantees to a degree the validity of the test. 
In order to determine the degree of validity, the test scores 
were correlated with teachers* ratings of the following six 
groups : 

(a) 78 pupils of the j f e * en f grades, ranked by two teachers; 





12 year Girls 


12 year Boys 


I. interest 


.871 


.828 


S. " 


.817 


.873 


A. " 


.483 


.867 


I. interest 


8.09 


8.52 


S. " 


5.88 


6.16 


A. " 


3.62 


4.60 


I. vs. S. interest 


.683 


.797 


I. vs. A. " 


.286 


.347 


S. vs. A. " 


.321 


.532 


I. interest 


2.91 


3.54 


S. " 


2.52 


2.20 


A. " 


2.60 


1.68 



(b) 29 

(c) 21 

(d) 22 

(e) 32 

(f) 24 



seventh grade, 



" the same 

teacher with an interval of 
a year between the ratings. 



The validity of the test as a measure of the three aspects 
of interest is indicated by the last column of (corrected) 
coefficients given in Table 174. The corrected coefficients 
were obtained by dividing the square root of the raw corre- 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 469 

lation between test and criterion by the square root of the 
reliability coefficient of the criterion. 

TABLE 174 

COEFFICIENTS SHOWING RELIABILITY OF CRITERION AND VALIDITY OF 

INTEREST TEST 



(a) 



(b) 



(c) 



(d) 



(e) 



(f) 



(The raw correlation given in each case is the higher one of the two.) 

The average corrected correlations with criterion for the 
six groups of children are as follows : 

For intellectual interest .65 

For social interest .496 

For activity interest .31 



Reliability 
of criterion 




Raw Correl. 
with Criterion 


Corrected Correl. 
with. Criterion 


]/ Rel. Coeff. 


f i 


.617 


.786 


.541 


.68 


I 


i S 


.365 


.604 


.245 


.41 


S 


IA 


.402 


.634 


.201 


.32 


A 


f 1 


.588 


.767 


.538 


.70 


I 


P 


.602 


.776 


.347 


.45 


S 


(A 


.292 


.540 


.376 


.70 


A 


f 1 


.767 


.876 


.696 


.79 


I 


S 


.709 


.842 


.574 


.68 


S 


IA 


.281 


.530 


-.072 




A 


f 1 


.598 


.773 


.391 


.51 


I 


P 


.570 


.755 


.392 


.52 


S 


IA 


.360 


.600 


.189 


.31 


A 


f 1 


.520 


.721 


.460 


.64 


I 


S 


.590 


.770 


.260 


.34 


S 


IA 


.465 


.682 


-.154 




A 


f 1 


.858 


.926 


.538 


.58 


I 


S 


.359 


.599 


.347 


.58 


S 


A 


.503 


.709 


.376 


.53 


A 



470 TRAITS OF GIFTED CHILDREN 

The first two of these are nearly as high as similar corre- 
lations yielded by the intelligence tests available a few years 
ago. The third is lower. All are high enough to make possi- 
ble a reasonably valid comparison of two such large groups 
of children as our gifted and control groups. 

CASE COMPARISONS 

Additional evidence as to validity was obtained by exam- 
ining the supplementary information furnished by parents 
and teachers concerning 32 gifted children who had earned 
high scores on at least one of the three kinds of interest. 

The first 16 cases are selected for high scores in intellec- 
tual interest (140 or above, as compared with norm of 118). 
Four of these also have high scores in social interest (130 or 
more, as compared with norm of 117), and one has a high 
score in activity interest (above 130, as compared with norm 
of 122). High scores are indicated in italics. In each case 
the first three figures give, in order, the scores on intellectual 
interest, social interest, and activity interest. 

Case I. S. A. 

(1) 145. 130. 125, girl 12, 145. Precocious charmingly 

natural. 

(2) 144. 135. 124, girl 11, 149. Likes history and geography 

musical genius extremely popular with 
adults and children alike likes to reason 
things out inventive turn of mind. 

(3) 143. 128. 130, hoy 12, 151. Strong in literary subjects 

wants to do newspaper work musical 
appreciation a "real" hoy, normal in all 
respects. 

(4) 142. 131. 122, girl 12, 142. Good literary taste whole 

family omnivorous readers likes reading, 
geography, arithmetic and history en- 
joys everything at school "a most inter- 
esting and delightful child," 

(5) 142. 130. 128, hoy 13, 146. Mathematical lively keen 

humour. 

(6) 142. 129. 124, girl 10, 149. Literary prefers reading 

tends to follow, not lead modest. 

(7) 142. 127. 127, boy 13, 142. Would rather read than play 

likes history no leadership dramatic 
ability enjoys music. 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 471 

Case ISA 

(8) U2. 122. 122, girl 12, 137. Good literary taste men- 

tally independent old for her age re- 
served should have more social contacts. 

(9) U2. 116. 128, girl 13, 130. Literary prefers reading- 

literary and artistic bent quiet shy. 

(10) Ul. 129. 127, girl 13, 149. Likes history wants to ex- 

plore and be an archaeologist loves ani- 
mals considered by some sluggish and 
inert. 

(11) Ul. 129. 125, boy 13, 136. Prefers reading quiet- 

must be leader or will not play in group 
does not make friends easily very in- 
terested in nature, camping, fishing, etc. 

(12) 141. 129. 123, boy 12, 155. Likes literature, composi- 

tion, history remarkable vocabulary a 
born leader. 

(13) Ul. 129. 122, boy 12, 140. Reads a great deal likes 

arithmetic prefers school to anything 
else diplomatic interested in people 
normal social activities a good swimmer 
interested in engineering. 

(14) IbQ. 129. 127, girl 13. Reads everything she can get 

most interested in languages main in- 
terests are novels and a "good time." 

(15) UQ. 126. 124, girl 11, 159. Prefers reading likes En- 

glish and music "needs more play." 

(16) 140. 118. 118, girl 12, 142. Interested in reading fond 

of music does not make necessary social 
contacts ranks average or below in all 
traits needs to read less and play more. 

Four of the above cases and eleven of the following have 
scores of 130, or higher, in social interest. One of the follow- 
ing has a score of 130 in activity interest: 



Case ISA 

(17) 138. 139. 129, girl 11, 146. (Highest average score for 

all three interests.) Musical ability 
enjoys school happy with other pupils 
can do every activity of child life a 
good mimic inordinately fond of ani- 
mals. 

(2) 144. 135. 124, See previous list. 



472 TRAITS OF GIFTED CHILDREN 

Case ISA 

(18) 136. 135. 130, girl 11, 158. Good all-round development 

musical talent prefers dancing natu- 
rally friendly manners extraordinary 
appreciation of beauty aesthetic tenden- 
cy keen humour. 

(19) 139. 134. 122, girl 12. Musical genius language apti- 

tude exceedingly attractive tends to be 
a leader rated as average for popularity 
only average physical energy. 

(20) 138. 133. 123, boy 13, 145. Reads a great deal likes 

arithmetic and literature popular keen 
sense of humour enjoys all outdoor 
sports. 

(21) 128. 133. 114, boy 12, 156. Likes history and science 

mathematical ability reads tremen- 
dously strong mechanical bent very 
nearsighted generally considered non- 
social but considered highly social by 
Dr. Terman's research assistant. 

(22) 135. 132. 127, girl 10, 145. Most interested in reading 

and history strong sense of beauty 
play interests normal but is handicapped 
in having only one child near for playing 
with. 

(23) 138. 131. 127, boy 12, 148. A reader likes arithmetic 

literature, science artistic in temper- 
ament has more than average ability in 
music and art likes baseball has no 
mechanical skill. 
(6) 142. 131. 122, See previous list. 

(24) 136. 131. 125, boy 12, 154. Interested in history, and 

geography, languages, arithmetic unas- 
suming but confident never quarrels 
and is never quarrelled with never 
bullies and is not easily bullied a boy of 
"genuine worth." 

(25) 134. 131. 124, girl 11, 147. Interested in literature- 

dreamy indefinite lacks curiosity 
keen musical appreciation "household 
revolves round her.'* 

(1) 145. 130. 125, See previous list. 

(5) 142. 130. 128, See previous list. 

(26) 137. 130. 126, girl 12, 150. Loves outside loves pets 

at present absorbed in art. 

(27) 134. 130. 120, girl 11, 149. Logical mind artistic incli- 

nation poise "domineered by mother." 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 473 

Two from the above lists and the following five cases 
have scores of 130, or higher in activity interest. 

Case ISA 

(28) 133. 125. 134, boy 11, 151. Interested in history prac- 

tical turn of mind very quick reactions. 

(29) 133. 126. 132, girl 11, 143. "Considerable pep" un- 

usual poise charming decidedly a 
leader unusual ability in directing class 
dramas, etc. great ability and tact in 
managing others extremely fond of 
music. 

(30) 132. 126. 131, boy 12, 148. (Little information avail- 

able a foreigner.) Interested in work 
and in play. 

(31) 133. 124. 131, girl 11, 132. Musical ability quiet and 

unassuming extraordinary amount of 
physical energy. 

(18) 136. 135. 130, See previous list. 
(3) 143. 128. 130, See previous list. 

(32) 129. 128. 130, girl 12, 140. Good story writer good in 

dramatics good in music and art 
leader in her work and in her play. 

From an examination of these cases, it is apparent that 
there is considerable agreement between the test results and 
the information supplied by parents and others. In no in- 
stance is there any marked discrepancy between the two 
sources of information; but in cases 10, 20, 21, and 25 there 
seems to be some contradiction. Number 10 is a child labeled 
"sluggish and inert" with a score of 127 in activity; but 
since the child is anxious to "travel and explore" the test 
score may give a better indication of her activity interest 
than does her behavior. Number 20 has only the average 
score for activity interest, but enjoys outdoor sports. Number 
21 is an interesting case. In the first place he reads enor- 
mously; but his intellectual interest score is lower than the 
average for gifted children. In the second place he is gener- 
ally regarded as non-social, whereas the test score ranks him 
exceptionally high in social interest. In the opinion of Dr. 
Terman's research assistant, who knows this boy well, he 
ranks very high in social interest. The test here may be 
bringing out what is not superficially apparent. Number 25 



474 TRAITS OF GIFTED CHILDREN 

has a high score for social interest, though she is character- 
ized as "dreamy, indefinite, lacking in curiosity." The ex- 
planation may be here in the statement, "the household re- 
volves round her." 

The agreement between the test results and the different 
reports on these 32 gifted children is then a further indica- 
tion that the test is a valid one for determining the direction 
of a child's interests. 

The above brief case studies also throw light on the type 
of child who is intellectually interested, on the type who is 
socially interested, and on the type who is interested in ac- 
tivity. As we should expect, the intellectually interested 
type has literary tastes, is fond of reading, enjoys history, 
geography, and mathematics. The social type may be popu- 
lar, but is not necessarily so. This type is happy with others, 
has a sense of humor and seems essentially to have aesthetic 
appreciation and the artistic temperament. The real activity 
type is not necessarily the one fond of outdoor games rather 
is it the born leader, the one who can manage Bothers, the 
one quick to respond; more briefly, the one with what is 
commonly called "pep." 

COMPARISON OF GIFTED AND CONTROL GROUPS 

Table 175 gives by age and sex the means and S.D's of 
the gifted and control group 1 on the three kinds of interest. 
The numbers are as follows : 

Boys Girls 



Age 


Gifted 


Control 


Gifted 


Centre 


8 


17 


_ 1 


20 


- 


9 


48 


_ 


43 


- 


10 


50 


43 


41 


47 


11 


75 


65 


59 


46 


12 


62 


54 


64 


68 


13 


61 


67 


50 


84 


14 


32 


47 


32 


42 


15 


20 


28 


15 


18 


Total 


365 


304 


324 


305 


Group total 


669 




629 





a The lowest grade tested for the establishment of norms was grade 4, 
as the test is not easily administered to unselected children below this level 
of school advancement. Control children below age 10 were omitted from 
the calculations. 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 475 



TABLE 175 

MEANS AND S.D's OF GIFTED AND CONTROL GROUPS ON INTELLECTUAL, 
SOCIAL, AND ACTIVITY INTERESTS 

Intellectual Interests 

Boys Girls 

Gifted Control Gifted Control 

Age M S.D. M S.D. M S.D. M S.D. 

8 

9 

10 
11 
12 
13 
14 
15 

Social Interests 

8 

9 

10 
11 
12 
13 
14 
15 

Activity Interests 



112.5 


10.27 


_ 


_ 


125.3 


7.64 


_ 





126.6 


7.50 


110.1 


11.36 


129.4 


5.80 


115.1 


10.26 


133.2 


5.11 


119.3 


9.79 


131.9 


6.18 


119.2 


9.21 


132.2 


5.20 


*119.0 


10.33 


129.3 


8.74 


*116.1 


8.35 



119.8 


5.65 





- 


124.7 


8.71 








128.3 


6.80 


112.3 


9.94 


130.3 


5.56 


116.0 


9.90 


131.4 


6.50 


118.8 


10.03 


130.0 


6.63 


120.3 


9.35 


133.0 


4.70 


*121.1 


7.52 


132.5 


4.84 


116.7 


9.37 



113.8 


5.67 


_ 


120.5 


4.15 


- 


120.6 


4.89 


111.6 


120.9 


4.60 


114.5 


123.1 


4.41 


116.8 


122.7 


4.95 


116.1 


122.4 


4.99 


M16.2 


121.4 


4.23 


*116.9 



6.04 
6.22 
6.69 
6.71 
6.47 
4.38 



118.4 


3.23 





122.3 


5.26 





122.8 


5.95 


. 115.0 


124.0 


4.99 


117.3 


124.0 


4.65 


117.4 


122.1 


4.25 


119.0 


123.9 


2.59 


M19.3 


123.9 


2.70 


*116.4 



6.86 
6.83 
6.67 
5.88 
5.76 
5.17 



8 
9 

10 
11 
12 
13 
14 
15 



121.8 
123.8 
124.3 
123.6 
123.3 
124.0 
122.9 
122.0 



3,98 
3.43 
3.28 
3.78 
3.59 
3.10 
3.06 
3.77 



120.0 

122.0 

123.4 

123.4 

*123.9 

*124.6 



4.03 
5.21 
4.32 
4.27 
4.77 
4.89 



122.6 
123.6 
122.6 
123.2 
122.3 
122.1 
121.8 
121.7 



2.70 
3.30 
4.35 
4.10 
3.76 
3.66 
3.39 
3.75 



120.7 

122.4 

121.8 

121.4 

*123.2 

*121.6 



3.24 
4.44 
4.10 
4.36 
3.78 
2.59 



' "Selected group" 14 and 15 year olds in high school not tested. 



From the above figures it is apparent 

(1) That there is considerable growth from year to year 
in intellectual interests, only a little in social interests, and 
almost none in activity interests; 

(2) That sex differences are non-existent for intellectual 
interests, but that girls rate slightly higher at nearly all ages 
in social interests, and boys slightly higher at most ages in 
activity interests; 

(3) That the means for the gifted are in most cases sig- 
nificantly higher than for the control. 

How significant the differences are between the gifted and 
control groups may be seen from Table 176, which gives 



476 



TRAITS OF GIFTED CHILDREN 



for the ages 10 to 13 the ratios of the differences to the PJE's 
of the differences. Wherever this ratio is greater than 3 P.E. 
the difference may be considered significant. 

TABLE 176 

SIGNIFICANCE OF THE DIFFERENCES BETWEEN GIFTED AND CONTROL 
GROUPS IN INTEREST SCORES 



Intellectual Interests 

Ratio 
Diff. in 


Social Interests 

Ratio 
Diff. i n 


Activity 
Diff. 


Interests 

Ratio 
in 
P.E's 

8.2 

3.5 

3.0 
1.4 
-0.35 
1.1 
1.36 
1.3 


Age 

10 
11 

12 
13 


Sex : 

Boys 
Girls 
Boys 
Girls 
Boys 
Girls 
Boys 
Girls 


P.E. of Diff. 

16.48 


P.E's 

12.0 
13.2 
14.7 
13.0 
13,9 
12,9 
13.6 
10.4 


P.E. of Diff. 

9.01 


P.E's P.E. of Diff 

11.6 4.30 


1.37 

16.0 
1.21 

14.29 

14.30 
13.84 


.78 
7.81 


8.5 
10.2 
8.27 
8.70 
9.85 
9.50 
5.15 


.52 
1.97 


.92 
6.42 


.56 
1.60 


.63 
6.70 

6.24 
"^72 

6.60 


.53 
.79 


.57 
-.13 


.998 
12.60 


.37 
.52 


.98 
12.65 


.67 
6.67 


.46 
.60 


.93 
9.81 


.70 
3.04 


.44 

.61 


.94 


.59 


.47 



The differences in the case of intellectual and social inter- 
ests are all highly significant, while those for activity inter- 
ests are significant only at age 10 for girls and ages 10 and 11 
for boys. The highest individual scores found were as fol- 
lows: 

For intellectual interests, 145, 12 year old gifted girl 
For social interests, 139, 11 year old gifted girl 
For activity interests, 138, 12 year old control boy 
For total of the three interests, 406, 11 year old gifted girl 

Figure 27, which gives the curves of mean score for the 
gifted and control groups, by age and sex, makes clear the 
relative positions of the two groups in each of the three 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 477 



FIGURE 27 
MEAN SCORES IN INTEREST TESTS, GIFTED AND CONTROL GROUPS, BY AGE 



155 



130- 
125; 

115- 
110 
135- 
130- 
125- 
JZO- 
115- 

135 
130 
125 
120- 

ItO 



Intellectual Interests 



BOYS 




GIRLS 




10 11 \Z 13 14 15 B 10 11 1 15 14 15 



Social Interests 



BOYS 




GIRLS 



3 10 11 \Z 13 14 15 9 10 11 12 13 14 15 



Activity Interests 



BOYS 



GIRLS 



9 10 11 \t 13 14 15 9 10 11 12 13 14 15 



478 TRAITS OF GIFTED CHILDREN 

types of interest. In contrast to the girls, both gifted and 
control boys show a decrease in intellectual interest from 12 
to 15 years. In activity interest the gifted, both girls and 
boys, show a noticeable tendency to decrease of score with 
age; with control boys the curve shows a small continuous 
rise, while with control girls it is approximately a plateau. 

That intellectually gifted children should surpass the 
average in their intellectual interests was to be expected. 
For the first time, however, we are able to express this supe- 
riority in quantitative terms; it is approximately 1.4 times 
the standard deviation found for unselected children of cor- 
responding age, or approximately half as great as their supe- 
riority in intellectual ability as measured by the Stanford- 
Binet test. In intellectual interest about 90 per cent of the 
gifted equal or exceed the mean of unselected children. The 
quantitative scores for the social and activity interests of 
gifted children add more to our previous knowledge than 
those for intellectual interests. Instead of lacking social in- 
terests, as many have believed, the typical gifted child stands 
in this respect about one standard deviation above the mean 
for unselected children of corresponding age. Roughly, 85 
per cent of the former equal or exceed the mean of the latter 
in social interest. In activity interests the gifted are neither 
more nor less than normal. Even if our test of interests 
should prove to be much too unreliable for use as a measure 
of the individual child, these conclusions would still be jus- 
tified in comparing a large group of gifted children with a 
large unselected group, and this is the point of chief interest. 

Evidence from other sources regarding the interests of 
our gifted subjects agrees in general fairly well with the 
evidence of the interest tests. These children have more 
than average liking for their school work (Chapter X) 
and for intellectual plays and games (Chapter XIV). They 
show somewhat less preference for active games than do 
average children (Chapter XIV) . Teachers and parents both 
rate them highest in such traits as originality, desire to know, 
and general intelligence. They rate them considerably lower 
on social traits like "fondness for large group," "leadership," 
and "popularity with other children," but on these traits, 
also, they rate them above unselected children (Chapter 
XVIII). Interest in activity, unfortunately, was not included 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 479 

in the 25 traits on which ratings were secured from parents 
and teachers. 

THE INFLUENCE OF INTEREST UPON SCHOOL ACHIEVEMENT 

Information on the relation between interest and achieve- 
ment was obtained by giving to a group of 81 children of the 
high sixth, low seventh, and high seventh grades the interest 
test, the National Intelligence Test, and the Stanford Achieve- 
ment Test. For this group the reliabilities of the tests, their 
raw intercorrelations, and their intercorrelations corrected 
for attenuation were found. Partial correlations were then 
computed, based on the corrected intercorrelations, to show 
the effect of eliminating, successively, intellectual interest, 
intelligence, social interest, and activity interest. For this 
group reliability coefficients were as follows : 

Intellectual Interest .935 

Social Interest .832 

Activity Interest .859 

National Intelligence Test .907 

Stanford Achievement Total .935 

The intercorrelations, corrected for attenuation, were as 
follows : 

Intel. Soc. Activ. 

Int. Int. Int. N.I.T. 

Social Interest .798 

Activity Interest .579 .512 

NaH Intel. Test .463 .498 .465 

Achievement Total* .629 .496 .397 .817 

* Intercorrelations for the separate tests of the Stanford Achievement battery 
are given in Table 119. 

From the above figures it is apparent (1) that each test 
measures something with a fairly high degree of reliability 
and (2) that this something is far from identical in the vari- 
ous tests. 

With intellectual interest rendered constant we have : 

Soc. Activ. 

Int. Int. N.I.T. 

Activity Interest .102 

National Intelligence Test .242 .272 

Achievement Total .011 .052 .763 



480 TRAITS OF GIFTED CHILDREN 

With intelligence (N J.T. score) rendered constant : 

Intel. Soc. Activ. 

Int. Int. Int. 

Social Interest .737 

Activity Interest .464 .365 

Achievement Total .490 .178 .033 

With social interest constant: 

Intel. Activ. 

Int. Int. N.I.T. 

Activity Interest .328 

National Intelligence Test .126 .250 

Achievement Total .445 .157 .757 

With activity interest constant : 

Intel. Soc. 

Int. Int. N.LT. 

Social Interest .716 

National Intelligence Test .269 .341 

Achievement Total .535 .372 .778 

The correlation between intelligence and the achieve- 
ment total, when the effect of intellectual interest is elimi- 
nated, is .763. The correlation between intellectual interest 
and the achievement total, when the effect of intelligence is 
eliminated, is .490. These correlations give a comparative 
measure of the effect of these two functions on success as 
measured by achievement in school subjects. Intellectual in- 
terest is a very potent factor in determining achievement. 
But the question arises, must a child be interested in what 
he is doing in order to achieve success in it, or is it the ability 
to succeed that gives the interest? In which direction does 
the causal relation lie? We find that the most successful 
child is highly intelligent and highly interested. Some chil- 
dren who are not highly interested have succeeded, but they 
are highly intelligent. Again, some highly intelligent, but not 
highly interested, have not succeeded; and, finally, some 
with lower intelligence and not a high degree of success are 
highly interested. The answer to the question, then, is that a 
child must be interested to achieve success, the greater the 
interest and the higher the intelligence, the greater the suc- 
cess and not that ability to succeed produces the interest. 

With regard to social interest and activity interest, when 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 481 

the effect of intelligence is eliminated, social interest shows 
a very slight correlation with achievement; and activity in- 
terest has no influence at all. 

Analyzing the data further, in order to determine which 
school subjects are most influenced by intellectual interest, 
with the effect of intelligence removed, we find it has, prob- 
ably, the greatest influence on arithmetical reasoning, though 
the influence is significant for all subjects except spelling. 
What little correlation there is with spelling could easily be 
due to chance, so that intellectual interest has no influence on 
success in spelling. Social interest, judging the correlations 
in the light of their P.E's, has very little effect on any of the 
school subjects; and activity interest has no influence at all. 

When, on the other hand, we examine the correlations 
between social interest and the school subjects when the ef- 
fect of intellectual interest is eliminated, we find there is no 
correlation; and with arithmetical reasoning there is proba- 
bly a negative correlation. Between activity interest and 
success in the school subjects there is no correlation at all. 

We are in a position, then, for judging the precise in- 
fluence of the various interests on achievement in school 
subjects. 

Other significant facts are brought out by the partial cor- 
relations. There is the greater association between intellec- 
tual and social interest, between intellectual and activity in- 
terest, and between social and activity interest than between 
any one of these and intelligence. This throws light on the 
meaning of interest itself. Further light is thrown on it when 
we consider, on the one hand, the lack of correlation between 
social interest and history information, the low correlation 
between social interest and science information, and between 
social interest and all school achievement, and, on the other 
hand, the very high correlation between social interest and 
intellectual interest. This is a definite denial that interest is 
knowledge. Again, the low correlation between activity in- 
terest and achievement, and the high correlation between in- 
telligence and achievement, throw important light on the 
meaning of intelligence scores they are void of meaning in 
terms of activity. This was also borne out by the lower 
scores of the gifted children in activity interest, compared 
with their scores in intellectual and in social interests. 



482 TRAITS OF GIFTED CHILDREN 

SUMMARY 

1. A free association test has been devised for the meas- 
urement of three types of interest : intellectual interest, social 
interest, and activity interest. The reliability coefficients of 
the test, .80 to .90, compare favorably with those of current 
intelligence tests. The validity coefficients are somewhat 
lower, but are high enough to permit valid comparisons be- 
tween groups of subjects, if not between individuals. The 
average of the validity coefficients found is .65 for intellec- 
tual interest, .50 for social interest, and .31 for activity in- 
terest. Case studies furnish additional evidence of validity. 

2. The new interest test was given to 689 gifted children 
and to 609 composing a control group (Control Group D). 

3. The scores indicate that there is considerable increase 
of intellectual interest with age, only a little increase in the 
case of social interest, and none with activity interest. 

4. There are no sex differences in intellectual interest, 
but girls surpass boys at nearly all ages in social interest and 
boys surpass girls at most ages in activity interest. 

5. In intellectual interest the mean score of the gifted 
children at most ages exceeds the mean of unselected chil- 
dren of corresponding age by approximately 1.4 times the 
S.D. of the latter. This is approximately half as great a de- 
gree of superiority as obtains in the case of intelligence. 
Stated in another way, about 90 per ce$t of the gifted chil- 
dren equal or exceed the mean of unselected children in 
intellectual interest. 

6. The superiority of the gifted in social interest is some- 
what less, but is still very decisive, the mean of the gifted 
being about one S.D. above that of the control group. Rough- 
ly, 84 per cent of gifted equal or exceed the mean of unse- 
lected children in social interest. 

7. With respect to activity interest, gifted and control 
groups do not differ materially. 

8. A special group that was given the Interest Test was 
also given the National Intelligence Test and the Stanford 
Achievement Tests. Intercorrelations of all these tests were 
computed and the method of partial correlation was applied. 
This treatment made it possible to secure a measure of the 
relative effect of intelligence and of intellectual interest upon 



INTELLECTUAL, SOCIAL, AND ACTIVITY INTERESTS 483 

achievement. The correlation between intelligence and 
achievement when intellectual interest is rendered constant 
is .76. The correlation between intellectual interest and 
achievement when intelligence is rendered constant is .49. 

9. The influence of intellectual interest on achievement 
is significant for all the school subjects except spelling, but 
is greatest in the case of arithmetical reasoning. 

10. Similar treatment shows that social interest and ac- 
tivity interest have, of themselves, practically no effect upon 
school achievement. 



CHAPTER XVII 



TESTS OF CHARACTER AND PERSONALITY TRAITS 1 

Tests of character and personality traits are a relatively 
new development in psychology and are still on a much less 
satisfactory basis than tests of intelligence, information, or 
school achievement. However, in view of the great desira- 
bility of obtaining character data which would be more ob- 
jective than the trait ratings supplied by parents and teach- 
ers, and in view of the promising results which had been 
secured by Voelker, Cady, and other workers in this field, a 
comparison of gifted and unselected children on a battery of 
such tests seemed worth undertaking. 

Methods of approach which were considered included 
the free association method of Kent and Rosanoff, the Pres- 
sey method of testing emotionality, the Downey will-tem- 
perament tests, the Woodworth-Cady questionnaire-test of 
psychopathic tendencies, and the methods developed by 
Voelker and Cady for testing honesty and incorrigibility. The 
Kent-Rosanoff test would doubtless have yielded results of 
great interest, but it had to be ruled out because it is nol 
adapted to group testing. The Pressey method was not very 
suitable for young children and had not been sufficiently 
validated. The Downey will-temperament test was dropped 
from consideration because of the extremely low reliability 
and validity coefficients which it has generally yielded. The 
Voelker-Cady tests, on the other hand, were suitable for 
children, had been subjected to rigid statistical procedures, 
and had been demonstrated to have a fairly high degree of 
reliability and validity. Mr. A. S. Raubenheimer undertook 
the task of preparing a series of tests which would be diag- 
nostic of much the same traits Voelker and Cady had at- 
tempted to measure, and at the same time free from certain 
objectionable features of their tests. Before describing his 
experiment, a brief account will be given of the work of 
Voelker and Cady- 

Written with the assistance of A* S. Raubenheimer and Florence L. 
Goodenoxigh. 

485 



486 TRAITS OF GIFTED CHILDREN 

Voelker devised ten tests of trustworthiness, the total 
score of which, applied to boys, correlated .75 with the total 
of another similar series and, on an average, .60 with the 
judgments of teachers and scout leaders. 1 The traits tested 
were as follows: 

1. Willingness to accept undeserved credit (overstate- 
ment test) 

2. Suggestibility, as indicated by the M and N test m the 
Downey will-temperament series 

3. Willingness to accept help in solving puzzles after 
promising not to accept help 

4. Conscientiousness in returning borrowed property 

5. Dishonesty in accepting overchange 

6. Willingness to accept a tip for a trifling courtesy 

7. Trustworthiness in performing a routine task under 
temptation to neglect it (pushing a button every two minutes 
for ten minutes) 

8. Similar to Test 7; task is to cancel A's in a book con- 
taining extremely interesting pictures 

9. Willingness to "peep" when placed on one's honor to 
perform a task with the eyes closed 

10. Willingness to cheat in scoring one's own responses to 
a test 

The Voelker tests were all of the "performance" type; 
that is, the subject was confronted by a natural situation in 
which there was genuine temptation and his score depended 
upon how he met the situation. There can be no doubt that 
they were severe tests of actual trustworthiness. Some of his 
tests were unsuitable for our purpose because they could 
not be given to groups, others because they necessitate sub- 
jecting the child to so much temptation that their use was 
considered objectionable for the present purposes. 

Cady 2 set himself the task of finding tests which would 
yield a measure of the moral traits involved in the adjust- 
ment of children to the social and mandatory requirements 

^aul Frederick Voelker: The Function of Ideals and Attitudes in So- 
cial Education. Teachers' College, N. Y., 1921, p. 126. 

2 Vernon S. Cady: The Estimation of Juvenile Incorrigibilitu. Journal 
of Delinquency Monograph No. 2, 1923, p. 140. Whittier State School, 
Whittier, Calif. This study was made under the joint supervision of T. L. 
Kelley, L. M. Terman, and J. Harold Williams. 



CHARACTER AND PERSONALITY TRAITS 487 

of the school; that is, measures of incorrigibility. Some of 
the tests which he tried out, among others, were the follow- 
ing: 

1. A test of trustworthiness in following directions in a 
motor task (dotting circles or tracing mazes with the eyes 
shut) when there is considerable temptation and opportu- 
nity to cheat by "peeping." (Modification of Voelker's 
test.) 

2. A test of honesty in scoring one's own intelligence test 
blank in the face of temptation and opportunity to cheat by 
copying in the correct response from the key while scoring. 
(Voelker) 

3. A test of modesty and accuracy in statements about 
knowledge possessed. (Modification of Voelker's overstate- 
ment test.) 

4. A test of moral judgment, in which the subject indi- 
cates, by the figures 1, 2, .3, or 4, degrees of blame attaching 
to each of a number of moral traits. (Modification of Pres- 
sey's test.) 

5. A modification of the Woodworth questionnaire-test 
of tendencies likely to be associated with psychopathy or 
emotional instability. 

These tests, which were all suitable for mass use, were 
given to unselected boys, delinquent boys, and boys spe- 
cially selected on the basis of several teachers' ratings of 
boys representing the extremes of corrigibility found among 
12 to 14 year olds in the public schools of Fresno, California. 

The composite rating secured for the last group had a re- 
liability of .95 and furnished an exceptionally satisfactory 
criterion for judging the validity of the tests. 

Reliability and validity coefficients found for the tests, 
based upon "data from 150 boys of 12 to 14 years, were as 
follows : 

Reliability Validity 

1. Trustworthiness in motor task .74 .40 

2. Honesty in scoring own test .58 .19 

3. Overstatement .58 .41 

4. Moral judgment .38 .20 to .31 

5. Woodworth questionnaire .75 .36 to .42 

Total weighted score of above .75 .58 



488 TRAITS OF GIFTED CHILDREN 

These correlations are lower than are yielded by the best 
intelligence tests with groups of equal heterogeneity, but 
they are as high as the correlations given by many achieve- 
ment tests in common use. They demonstrate, rather con- 
clusively, the value of the test method in the study of char- 
acter traits. 

Raubenheimer arranged a new form of the overstate- 
ment test and devised a set of six new tests, all intended to 
throw light upon tendencies to moral instability. A brief de- 
scription of each follows: 

1. Overstatement, A. This test is a modification of one 
which had been used by Frangen. Fifty took titles were 
listed on the test blank, 20 of which were fictitious. The 
score is the number of fictitious titles marked. 

The instructions were as follows : 

"We want to see who has read the most books. You are to mark 
a cross on the dotted line in front of every hook you have ever read 
no matter how long ago you read it. After you have finished mark- 
ing the crosses, count up the number of crosses you have marked. 
This number will be your score. We want to see who will have the 
best score. Afterwards we will have you stand up and tell your 
score. A PERFECT SCORE IS 50. Look at the sample. 

Sample: 1. X Anderson's Fairy Tales." 
Fifteen of the fifty items follow : 

Robinson Crusoe 

Little Men 

Uncle Remus Stories 

The Underground Patrol 

White Fang 

Seaside Adventures 

Hans Brinker 

Scouting in Strange Lands 

Five Little Peppers 

By England's Aid 

Call of the Wild 

Campaigning in Argentina 

Tom Sawyer 

Sunk Without a Trace 

The Friday Murders 

Now count up your score and put the number you have read in 
the space here. Remember, a PERFECT SCORE is 50. 

MY SCORE IS 



CHARACTER AND PERSONALITY TRAITS 489 

2. Overstatement, B. This test, which is a modification of 
a test used by Voelker, consists of two parts. The first part 
calls for a statement of amount of knowledge the child has 
in regard to 80 items of information. The second part, given 
later without warning, tests the child's actual information 
on these same 80 items. The score is the per cent of over- 
statement or understatement. 1 

Directions: On the dotted line before each question put 2, 1, 
or to tell how well you know the thing it asks about. Put down 2 
if you know it VERY WELL. Put down 1 if you know it FAIRLY 
WELL. Put down if you KNOW NOTHING ABOUT IT. 

First sample: Can you ride a bicycle? 

Second sample: Can you skate on roller skates? 

Third sample: Can you drive a car? 

Sample items follow: 

Do you know who discovered America? 

Do you know who wrote "Huckleberry Finn 3 '? 

Do you know who was the prophet who spent the night in the lions' 
den? 

Do you know how to find the square root of decimals? 

Do you know how many degrees there are on a Centigrade thermom- 
eter? 

Do you know where Calcutta is? 

Do you know what causes an eclipse of the sun? 

Do you know what the receiving wires of a wireless are called? 

Do you know how water enters the roots of plants? 

Now add up your score. A PERFECT SCORE is 160 points. 



MY SCORE IS- 



In the second part the subject was told to underline the 
correct word in such sentences as the following : 

America was discovered by Drake Columbus Balboa Cook 
"Huckleberry Finn" was written by Alger Dickens Henty 

Mark Twain 
The prophet who spent the night in the lions' den was Daniel 

Jonah David Joel 

The square root of .0081 is .9 .09 .009 9 
The number of degrees on a Centigrade thermometer is 32 100 

180 212 

a See p. 505 for a description of the scoring method used with the gifted 
and control groups. 



490 TRAITS OF GIFTED CHILDREN 

Calcutta is in India Egypt Siberia Mexico 

An eclipse of the sun is caused by the shadow of the earth moon 
Mars Jupiter 

The receiving wires of a wireless are called the amplifiers de- 
tectors reflectors antennae 

Water enters the roots of plants by capillarity osmosis evapo- 
ration solution 

3. Questionable reading preferences. The child is asked 
to place a number from 1 to 10 before each of ten book titles 
to indicate how well he would like to read the book, 1 mean- 
ing greatest preference, 10 least. The score is the sum of the 
squares of the deviations of the individual items from the 
correct order as determined by competent judges. Two lists 
were given, separately. One of them is as follows : 



..A Daring Rescue 

..Roy Black, The Master Thief 

..Captains of Great Teams 

..Hobo Stories 

..Running Away with the Circus 

..The Adventures of Boys Who Became Great Men 

..Summer Camp Adventures 

..With the Gang in the Back Streets 

..The Boy Inventor 

..The Escape Through the Woods 



4. Questionable character preferences. This test consists 
of eight brief paragraphs, each characterizing a boy. The 
subject is asked to place before each characterization a num- 
ber from 1 to 8, to show how well he would like to have as a 
chum the boy described. The test is scored in the same man- 
ner as test 3. Two such lists of eight items each are given 
separately. Four of the items in one list are as follows : 

DICK joined the Boy Scouts as soon as he was old enough. He did 
not like it at first; the drill and the rules were hard. Now he is a 
troop leader and is planning a camp in the mountains next 
summer. 

RAY STEVENS is at school now, but he is anxious to get out. He wants 
to become a taxi-driver. Ray says that taxi-drivers have an easy 
time; they need not work so hard, and they go ahout a great deal. 

TED is a poor hoy; and, although he is at school, he must help provide 
for the family. Ted studies hard and also plays on the team; he 
wants to become a doctor. 



CHARACTER AND PERSONALITY TRAITS 491 

BILL EVANS is fourteen, and is the leader of his gang. He always 
manages to get his men home safely after they have had a good 
time around the pool room. Just last week they saw Tom Mix 
at the movies without paying. 

5. Social attitudes. Twenty-four things or ideas were 
named, each followed by four statements expressing various 
kinds of reactions to the thing or idea presented. The sub- 
ject is asked to check the one statement that most nearly 
tells how he feels about it. Score is the number of question- 
able items checked. Samples are as follows : 

CHUMS : It is hard to go without them. 

You cannot always trust them. 

They sometimes squeal on you. 

It is best to have them in your gang. 

BOY SCOUTS : They have too many rules. 

They have to drill too hard; it is no fun. 

They are regular fellows and have lots of fun. 

They are like sissies. 

TEACHERS : They work hard. 

They know they can punish you. 

They are not fair to you. 

They are kind of cranky. 

PLAYGROUNDS : There are always fellows watching you. 

They make you play games you don't like. 

There is no chance to do what you want to. 

You can have a good time there. 

POLICEMEN : They have it in for the kids. 

They are glad to help you out. 

It is fun to fool them. 

They are just hig bluffs. 

HAVING A PAPER 

ROUTE: It gives you a chance to get away from home. 

You can earn some money. 

You have a chance to get round town alone. 

You don't have to be so much with your 

lessons. 

6. Activity preferences. Of three things to do, the subject 
was to check the one thing he liked best. The test contained 
twelve items of this nature, the score being the number of 
items in which some other than the "best" response was 
checked. This test was not used with the gifted children for 
fear of possible criticism. It is hardly adapted for use with 
children of superior social and moral environment. Two of 
the items are as follows : 



492 TRAITS OF GIFTED CHILDREN 

A Go camping with the Boy Scouts. 

Go around seeing the country, getting lifts as you go. 

Quit school and go with the circus. 

B Match pennies and win. 

Have a paper route. 

Win money at the shoeting gallery. 

7. Rating the seriousness of offenses. This is a carefully 
worked out version of a test which has long been in use. Ten 
offenses were listed, covering a wide range in nature and 
seriousness. They had been selected from a much larger 
number of offenses actually committed by delinquent boys, 
and they differed from one another in seriousness by equal 
steps. The subject was to number them from 1 to 10 ac- 
cording to seriousness. The score was the sum of the squares 
of the deviations from the correct order. This test was fi- 
nally omitted from the series used with our gifted children; 
it correlated too highly with intelligence and some of its 
items were deemed objectionable on moral grounds. Some 
of the items were as follows : 

(b) Sam set fire to the public school which he attended. 

(d) Bob ran away from home and got a job, getting his room and 

board from another family, 
(f) Ted played hookey to go to a circus, 
(j) Joe entered the house of the people next door and took $2.50. 

The derivation of the above tests involved in each case a 
considerable amount of experimental work which we can- 
not here enter into. 1 The purpose of the tests was to find 
whether differences known to exist between two groups with 
respect to their social and moral adjustments can be detected 
in a test situation. After a preliminary experiment each test 
was made up in two forms and the entire battery was given 
to the following groups of subjects, all boys: 

1A. 50 boys in a superior type of school, each selected by 
the teachers as belonging in the highest quartile of 13 year 
old boys with respect to reliability, stability, and healthy- 
mindedness. 

The methods and results are set forth by Dr. Raubenheimer in the fol- 
lowing publication: An Experimental Study of Some Behavior Traits of 
the Potentially Delinquent Boy. The Psychological Monographs, 1925. 



CHARACTER AND PERSONALITY TRAITS 493 

IB. 37 boys, 13 years old, in the same school, selected by 
the teachers as belonging in the lowest quartile with respect 
to the above traits. 

2A. 42 boys, 11 years old, in the same school, correspond- 
ing to group Al. 

2B. 36 boys, 11 years old, in the same school, correspond- 
ing to group A2. 

3A. 43 boys, 13 years old, in the highest quartile in a 
school of somewhat inferior social status. 

3B. 40 boys, 13 years old, in the lowest quartile of this 
school. 

4A. 41 boys, 11 years old, in the highest quartile. 

5B. 36 boys, 11 years old, in the lowest quartile. 

6. 42 boys of 12 to 14 years in a parental school in the 
same city (delinquents). 

7. 36 boys of 13 to 14 years in the Whittier State School 
for delinquents. 

Both forms of the tests were given to all groups, and in 
addition, Form B of the National Intelligence Tests. The 13 
year olds of groups IB and 3B combined were used for de- 
termining the reliabilities of the tests. In the case of each 
of the character tests, one form was correlated with the 
other, and the reliability of the two forms combined was 
then computed by the use of Brown's formula. The resulting 
coefficients were as follows: 

Reliability 
Coefficients 

1. Overstatement A, books read 74 

2. Overstatement B, knowledge claimed 78 

3. Reading preferences 80 

4. Social preferences 79 

5. Social attitudes 77 

6. Activity preferences 74 

7. Offense ratings 78 

8. National Intelligence scores 86 

Table 177 gives for the same 13 year olds the intercor- 
relations of the tests, the intercorrelations corrected for at- 
tenuation (labeled C), and (in parentheses) the corrected 
intercorrelations found when intelligence is rendered con- 
stant by the method of partial correlation. 



494 TRAITS OF GIFTED CHILDREN 

TABLE 177 
INTERCORREIATIONS OF THE RAUBENHEIMER TESTS 



2. 
3. 
4. 
5. 


1 
Over- 
state- 
ment A 
Over- 
state- .45 (.39) 
ment B C.36 

Reading 
Pref- .30 
erence C.39 (.31) 

Social 
Pref- .14 
erence C.19 (.12) 

Social 


2 
Over- 
state- 
ment B 

.50 
.60 (.44) 

33 
'.40 (.32) 


3 
Read- 
ing 
Pref, 

.62 
.78 (.77) 


4 

Social 
Pref. 


5 6 7 
Social Activ- 
Atti- ity Offense 
tudes Pref. Ratings 




Atti- .19 




,38 




.41 




.35 










tudes C.25 


(.08) 


.46 


(.17) 


.52 


(.31) 


.45 (.39) 








6. 


Activity 






















Pref- .10 




.39 




.36 




.40 


.56 






erence C.13(-.06) 


.49 


(.05) 


.47 


(.26) 


.52 (.47) 


.74 


(.54) 




7. 


Offense 






















Rat- .25 




.50 




.27 




.10 


.49 


.23 






ings C.33 


(.19) 


.61 


(.08) 


.34 (-.03) 


.12 (-.01) 


.63 


(.17) .30 (- 


-.37) 


8. 


National 






















Intel- .22 




.64